%0 Edited Book %B Lecture Notes in Computer Science %C Bhubaneswar, India %D 2021 %E Diganta, Goswami %E Truong, Hoang %F SisLab:4607 %I Springer %P 309 %T ICDCIT: International Conference on Distributed Computing and Internet Technology %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4607/ %V 12582 %X This volume contains the papers selected for presentation at the 17th International Conference on Distributed Computing and Internet Technology (ICDCIT 2021) held during January 7–10, 2021, in Bhubaneswar, India. %0 Edited Book %B Studies in Computational Intelligence book series (SCI, volume 769) %E Sieminski, Andrzej %E Kozierkiewicz, Adrianna %E Nunez, Manuel %E Ha, Quang Thuy %F SisLab:2938 %I Springer, Cham %T Modern Approaches for Intelligent Information and Database Systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2938/ %V 769 %X Introduction This book offers a unique blend of reports on both theoretical models and their applications in the area of Intelligent Information and Database Systems. The reports cover a broad range of research topics, including advanced learning techniques, knowledge engineering, Natural Language Processing (NLP), decision support systems, Internet of things (IoT), computer vision, and tools and techniques for Intelligent Information Systems. They are extended versions of papers presented at the ACIIDS 2018 conference (10th Asian Conference on Intelligent Information and Database Systems), which was held in Dong Hoi City, Vietnam on 19–21 March 2018. What all researchers and students of computer science need is a state-of-the-art report on the latest trends in their respective areas of interest. Over the years, researchers have proposed increasingly complex theoretical models, which provide the theoretical basis for numerous applications. The applications, in turn, have a profound influence on virtually every aspect of human activities, while also allowing us to validate the underlying theoretical concepts. %0 Book %A Abdallah, Abderazek Ben %A Dang, Nam Khanh %D 2022 %F SisLab:4655 %I Springer International Publishing %T Neuromorphic Computing Principles and Organization %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4655/ %X his book focuses on neuromorphic computing principles and organization and how to build fault-tolerant scalable hardware for large and medium scale spiking neural networks with learning capabilities. In addition, the book describes in a comprehensive way the organization and how to design a spike-based neuromorphic system to perform network of spiking neurons communication, computing, and adaptive learning for emerging AI applications. The book begins with an overview of neuromorphic computing systems and explores the fundamental concepts of artificial neural networks. Next, we discuss artificial neurons and how they have evolved in their representation of biological neuronal dynamics. Afterward, we discuss implementing these neural networks in neuron models, storage technologies, inter-neuron communication networks, learning, and various design approaches. Then, comes the fundamental design principle to build an efficient neuromorphic system in hardware. The challenges that need to be solved toward building a spiking neural network architecture with many synapses are discussed. Learning in neuromorphic computing systems and the major emerging memory technologies that promise neuromorphic computing are then given. A particular chapter of this book is dedicated to the circuits and architectures used for communication in neuromorphic systems. In particular, the Network-on-Chip fabric is introduced for receiving and transmitting spikes following the Address Event Representation (AER) protocol and the memory accessing method. In addition, the interconnect design principle is covered to help understand the overall concept of on-chip and off-chip communication. Advanced on-chip interconnect technologies, including si-photonic three-dimensional interconnects and fault-tolerant routing algorithms, are also given. The book also covers the main threats of reliability and discusses several recovery methods for multicore neuromorphic systems. This is important for reliable processing in several embedded neuromorphic applications. A reconfigurable design approach that supports multiple target applications via dynamic reconfigurability, network topology independence, and network expandability is also described in the subsequent chapters. The book ends with a case study about a real hardware-software design of a reliable three-dimensional digital neuromorphic processor geared explicitly toward the 3D-ICs biological brain’s three-dimensional structure. The platform enables high integration density and slight spike delay of spiking networks and features a scalable design. We present methods for fault detection and recovery in a neuromorphic system as well. Neuromorphic Computing Principles and Organization is an excellent resource for researchers, scientists, graduate students, and hardware-software engineers dealing with the ever-increasing demands on fault-tolerance, scalability, and low power consumption. It is also an excellent resource for teaching advanced undergraduate and graduate students about the fundamentals concepts, organization, and actual hardware-software design of reliable neuromorphic systems with learning and fault-tolerance capabilities. %0 Journal Article %A Abdallah, Abderazek Ben %A Dang, Nam Khanh %F SisLab:4456 %I Frontiers %J Frontiers in Neuroscience %T Towards Robust Cognitive 3D Brain-inspired Cross-paradigm System %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4456/ %X Spiking Neuromorphic systems have been introduced as promising platforms for energy-efficient spiking neural network (SNNs) execution. SNNs incorporate neuronal and synaptic states in addition to the variant time scale into their computational model. Since each neuron in these networks is connected to many others, high bandwidth is required. Moreover, since the spike times are used to encode information in SNN, a precise communication latency is also needed, although SNN is tolerant to the spike delay variation in some limits when it is seen as a whole. The two-dimensional packet-switched network-on-chip was proposed as a solution to provide a scalable interconnect fabric in large-scale spike-based neural networks. The 3D-ICs have also attracted a lot of attention as a potential solution to resolve the interconnect bottleneck. Combining these two emerging technologies provides a new horizon for IC design to satisfy the high requirements of low power and small footprint in emerging AI applications. Moreover, although fault-tolerance is a natural feature of biological systems, integrating many computation and memory units into neuromorphic chips confronts the reliability issue, where a defective part can affect the overall system's performance. This paper presents R-NASH - a reliable three-dimensional digital neuromorphic system geared explicitly toward the 3D-ICs biological brain's three-dimensional structure, where information in the network is represented by sparse patterns of spike timing and learning is based on the local spike-timing-dependent plasticity rule. Our platform enables high integration density and small spike delay of spiking networks and features a scalable design. R-NASH is a design based on the Through-Silicon-Via technology, facilitating spiking neural network implementation on clustered neurons based on Network-on-Chip. We provide a memory interface with the host CPU, allowing for online training and inference of spiking neural networks. Moreover, R-NASH supports fault recovery with graceful performance degradation. %0 Journal Article %A Abdallah, Abderazek Ben %A Dang, Nam Khanh %D 2021 %F SisLab:4459 %I Frontiers %J Frontiers in Neuroscience %T Towards Robust Cognitive 3D Brain-inspired Cross-paradigm System %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4459/ %X Spiking Neuromorphic systems have been introduced as promising platforms for energy-efficient spiking neural network (SNNs) execution. SNNs incorporate neuronal and synaptic states in addition to the variant time scale into their computational model. Since each neuron in these networks is connected to many others, high bandwidth is required. Moreover, since the spike times are used to encode information in SNN, a precise communication latency is also needed, although SNN is tolerant to the spike delay variation in some limits when it is seen as a whole. The two-dimensional packet-switched network-on-chip was proposed as a solution to provide a scalable interconnect fabric in large-scale spike-based neural networks. The 3D-ICs have also attracted a lot of attention as a potential solution to resolve the interconnect bottleneck. Combining these two emerging technologies provides a new horizon for IC design to satisfy the high requirements of low power and small footprint in emerging AI applications. Moreover, although fault-tolerance is a natural feature of biological systems, integrating many computation and memory units into neuromorphic chips confronts the reliability issue, where a defective part can affect the overall system's performance. This paper presents R-NASH - a reliable three-dimensional digital neuromorphic system geared explicitly toward the 3D-ICs biological brain's three-dimensional structure, where information in the network is represented by sparse patterns of spike timing and learning is based on the local spike-timing-dependent plasticity rule. Our platform enables high integration density and small spike delay of spiking networks and features a scalable design. R-NASH is a design based on the Through-Silicon-Via technology, facilitating spiking neural network implementation on clustered neurons based on Network-on-Chip. We provide a memory interface with the host CPU, allowing for online training and inference of spiking neural networks. Moreover, R-NASH supports fault recovery with graceful performance degradation. %0 Conference Paper %A Abdelmoutia, Telli %A Ma, Thi Chau %A Mustapha, Bourahla %A Karim, Tabia %A Salem, Benferhat %B The International Conference on Pattern Recognition and Artificial Intelligence %C Montréal, Canada %D 2018 %F SisLab:3045 %T AniAge Ontology for Movement Classification in Vietnamese Dance %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3045/ %X This paper proposes an OWL ontology called \AniAge", to define taxonomy of dance movement classes and their relationships for the traditional Vietnamese dance taking into account the semantics of its art and its cultural anthropologists. The \AniAge" terminology can be used to describe elementary movements (poses) as a dataset ontology importing \AniAge". These poses are results of dance sequences segmentation (using segmentation techniques). The ontolgy \AniAge" is supported by classification rules, which are developed with the OWL complementary language SWRL (Semantic Web Rule Language) to entail movement phrases, which are basic movements with complete meaning. The dataset ontology containing poses descriptions can be queried using the query language SQWRL (Semantic Query Webenhanced Rule Language), which is extension of SWRL to retrieve implicit dance knowledge. Then, the query answers can be used for computer animation. %0 Book Section %A Bourahla, Mustapha %A Telli, Abdelmoutia %A Benferhat, Salem %A Ma, Thi Chau %B Digital Human Modeling and Applications in Health, Safety, Ergonomics and Risk Management: Modelling the Human Body and Motion %D 2019 %F SisLab:3524 %I Springer Nature Switzerland AG %T Classifying Non-elementary Movements in Vietnamese Mo Dances %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3524/ %X This paper proposes a method to classify non-elementary movements in Vietnamese dances. This classification method uses an OWL ontology called VDM (Vietnamese Dance Movements) recently developed by the authors. The VDM defines a taxonomy of dance move- ment classes and their relationships for the traditional Vietnamese dances taking into account the semantics of its art and its cultural anthropol- ogists. The VDM terminology describes elementary movements (poses) as a dataset ontology importing the ontology VDM. These poses are results of dance sequences segmentation (using segmentation techniques). In this paper, we support the initial ontology VDM by complex classification rules written with SWRL (Semantic Web Rule Language, which is the OWL complementary language) to classify non-elementary movements. The objective is to entail classes of movement phrases, which are non-elementary basic movements with complete meaning and illustrated using M ̃o dances. The classification result is the initial dataset VDM ontology augmented with class descriptions of non-elementary movements, which can be queried using the query language SQWRL (Semantic Query Web-enhanced Rule Language). %0 Conference Paper %A Bui, Anh Tu %A Le, Hong Anh %A Truong, Ninh Thuan %B 6th EAI International Conference on Context-Aware Systems and Applications (ICCASA) %C Quy Nhon %D 2017 %F SisLab:2657 %T Generation of Power State Machine for Android devices %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2657/ %0 Conference Paper %A Bui, Anh Tu %A Nguyen, Van Viet %A Truong, Ninh Thuan %B 2020 7th NAFOSTED Conference on Information and Computer Science (NICS) %D 2020 %F SisLab:4097 %T Generating test data for energy property in mobile applications %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4097/ %X In recent years, mobile software has been rapidly developing. The key requirement of mobile software is to ensure optimum power throughout its run. Unlike testing other software properties, energy testing is usually done on real devices to check actual energy consumption. As such, testers need to have optimal test data to save time and effort in testing. In this paper, we propose an approach to automatically generating test data in mobile software testing, concentrate on energy properties. The approach, firstly, builds CFGs (Control Flow Graph) from the source code of the application. We then propose algorithms to select testable paths from CFG such that the mobile applications consume more energy. Finally, test data are generated from the testable paths. %0 Conference Paper %A Bui, Anh Tu %A Truong, Ninh Thuan %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1517 %T Kiểm chứng phần mềm dựa trên phân tích chương trình %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1517/ %X Việc phát triển phần mềm yêu cầu các bước thực hiện chặt chẽ, độ chính xác cao nhằm đáp ứng đúng, đủ yêu cầu người sử dụng. Tuy nhiên đội ngũ phát triển phần mềm cũng không thể kiểm soát hết các lỗi, các vấn đề phát sinh khi phần mềm đưa vào sử dụng. Đề tài tập trung vào việc nghiên cứu các phương pháp nhằm phân tích một số thuộc tính của chương trình, từ đó xây dựng các công cụ phân tích chương trình từ nhiều khía cạnh khác nhau, đưa ra thông tin cho người phát triển phần mềm đảm bảo tính đúng đắn cho chương trình. %0 Conference Paper %A Bui, Ba Nam %A Phan, Anh %A Nguyen, Thi Nhat Thanh %B 2020 7th NAFOSTED Conference on Information and Computer Science (NICS) %F SisLab:4349 %T Land-Cover Mapping from Sentinel Time-Series Imagery on the Google Earth Engine: A Case Study for Hanoi %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4349/ %X Over the past decade, satellite image processing is an overwhelming bulk of work. Recently, with rapid development in information technology, Google released Google Earth Engine (GEE), which is a powerful cloud computing platform, to help to improve the performance of geospatial big data archives and processing. In this study, we deployed a machine learning model to evaluate the capability of time series Sentinel imagery (Sentinel 2 A/B and Sentinel 1A) in landcover mapping for Hanoi in 2019. First, we evaluated several traditional machine learning models, as a result, XGBoost classifier stands out as the best model with 86% overall accuracy (OA). As Hanoi is a frequent cloud-covered area, the combination of optical data and radar data helps to improve the quality of the landcover map in 2019. The use of GEE has made it easier and faster through the provided JavaScript API when ensuring high accuracy. %0 Conference Paper %A Bui, Dinh Luyen %A Nguyen, Tri Thanh %A Ha, Quang Thuy %B 2014: the 2nd International Conference on Computer Science, Applied Mathematics and Applications (ICCSAMA) %C Budapest, Hungary %D 2014 %F SisLab:1304 %P 313-324 %T Measuring the Influence of Bloggers in Their Community Based on the H-Index Family %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1304/ %V 282 %0 Conference Paper %A Bui, Duc Thinh %A Nguyen, Do Van %A Nguyen, Thi Thu Thuy %A Tran, Quoc Long %A Le, Thanh Ha %B (MIWAI 2018) 12th Multi-disciplinary International Conference on Artificial Intelligence %C Hanoi, Vietnam %D 2018 %F SisLab:3264 %T Aerial Image Semantic Segmentation Using Neural Search Network Architecture %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3264/ %X In remote sensing data analysis and computer vision, aerial image segmentation is a crucial research topic, which has many appli- cations in environmental and urban planning. Recently, deep learning is using to tackle many computer vision problem, including aerial image segmentation. Results have shown that deep learning gains much higher accuracy than other methods on many benchmark data sets. In this work, we propose a neural network called NASNet-FCN, which based on Fully Convolutional Network - a frame work for solving semantic segmenta- tion problem and image feature extractor derived from state-of-the-art object recognition network called Neural Search Network Architecture. Our networks are trained and judged by using benchmark dataset from ISPRS Vaihingen challenge. Results show that our methods achieved state-of-the-art accuracy with potential improvements. %0 Journal Article %@ 0973-7006 %A Bui, Hong Nhung %A Ha, Quang Thuy %A Nguyen, Tri Thanh %D 2017 %F SisLab:2817 %J Journal of Electronics and Communications %T A novel similarity measure for trace clustering based on normalized google distance %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2817/ %X In trace clustering, a problem of process mining, traditional distance measures only focus on the local relationship between trace pairs. In this paper, we propose a new method to measure the global relationship of the traces based on the Normalized Google Distance. Experimental results show that our method not only outperforms alternatives but also helps to speed up the trace clustering %0 Journal Article %@ ISSN: 0973-5763 %A Bui, Hong Nhung %A Ha, Quang Thuy %A Nguyen, Tri Thanh %D 2018 %F SisLab:3160 %I The Pushpa Publishing House %J JP Journal of Heat and Mass Transfer %N 3 %P 341-346 %T A novel similarity measure for trace clustering based on normalized google distance. %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3160/ %V Specia %X In trace clustering, a problem of process mining, traditional distance measures only focus on the local relationship between trace pairs. In this paper, we propose a new method to measure the global relationship of the traces based on the Normalized Google Distance. Experimental results show that our method not only outperforms alternatives but also helps to speed up the trace clustering. %0 Book Section %A Bui, Hong Nhung %A Nguyen, Tri Thanh %A Nguyen, Thi Cham %A Ha, Quang Thuy %B International Joint Conference on Rough Sets (IJCRS 2018) %D 2018 %E Nguyen, Hung Son %E Ha, Quang Thuy %E li, Tianrui %E Małgorzata, Przybyła-Kasperek %F SisLab:3159 %I Springer %N 11103 %P 644-657 %S Lecture Notes in Computer Science %T A New Trace Clustering Algorithm Based on Context in Process Mining %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3159/ %X In process mining, trace clustering is an important technique that at-tracts the attention of researchers to solve the large and complex volume of event logs. Traditional trace clustering often uses available data mining algorithms which do not exploit the characteristic of processes. In this study, we propose a new trace clustering algorithm, especially for the process mining, based on the using trace context. The proposed clustering algorithm can automatic detects the number of clusters, and it does not need a convergence iteration like traditional ones like K-means. The algorithm takes two loops over the input to generate the clusters, thus the complexity is greatly reduced. Experimental results show that our method also has good results when compared to traditional methods. %0 Conference Paper %A Bui, Hong Nhung %A Vu, Trong-Sinh %A Nguyen, Tri Thanh %A Nguyen, Thi Cham %A Ha, Quang Thuy %B The 11th IEEE International Conference on Knowledge and Systems Engineering %C Da Nang, Vietnam %D 2019 %F SisLab:3711 %T A Compact Trace Representation Using Deep Neural Networks for Process Mining %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3711/ %X — In process mining, trace representation has a significant effect on the process discovery problem, the challenge is to get highly informative but low-dimensional of the vector space from event logs. This is required to improve the quality of the trace clustering problem and generate the process models that are easy to understand. Though traditional trace representation methods have specific good effects, their vector space often has a lot of dimensions. In this paper, we address this problem by proposing a new trace representation method based on the deep neural networks. Experimental results show that our method not only outperforms alternatives but also helps significantly reduce the dimension of feature representation. %0 Conference Paper %A Bui, Hong-Nhung %A Vu, Trong-Sinh %A Nguyen, Hien-Hanh %A Nguyen, Tri-Thanh %A Ha, Quang-Thuy %A Trường Đại học Công nghệ, ĐHQGHN, %A Học viện Ngân hàng, %B Asian Conference on Intelligent Information and Database Systems %C Phuket, Thailand %D 2020 %F SisLab:4042 %P 35-46 %T Exploiting CBOW and LSTM Models to Generate Trace Representation for Process Mining %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4042/ %V 1178 %X In the field of process mining, one of the challenges of the trace representation problem is to exploit a lot of potentially useful information within the traces while keeping a low dimension of the corresponding vector space. Motivated by the initial results of applying the deep neural networks for producing trace representation, in this paper, we continue to study and apply two more advanced models of deep learning, i.e., Continuous Bag of Words and Long short-term memory, for generating the trace representation. The experimental results have achieved significant improvement, i.e., not only showing the close relationship between the activities in a trace but also helping to reduce the dimension of trace representation. %0 Conference Paper %A Bui, Ngoc Dung %A Hoang, Xuan Tung %B The 5th International Conference on Next Generation Computing 2019 %C Chiang Mai, Thailand %D 2019 %F SisLab:3838 %T Multiple Vehicles Tracking in Intelligent Transportation System using Convolutional Neural Network and Kalman Filter %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3838/ %X Vehicles detection and tracking have become an important role to traffic management systems. Recently, many vehicles tracking approaches have already been proposed. However, these approaches were unable to adequately distinguish vehicles from each other when those vehicles look similar and involve in complex transportation conditions. In this paper, a method for tracking vehicles in surveillance cameras is presented. In our method, Convolutional Neural Networks is used to detect vehicles. Also, multiple Kalman filters are used to track those vehicles. The proposed method is designed for distinguishing and tracking multiple vehicles simultaneously. Our experiments show that the proposed mechanism achieves high accuracy even with real time constraints. %0 Conference Paper %A Bui, Ngoc Thang %B 2014: The 18th Pacific-Asia Conference on Knowledge Discovery and Data Mining (PAKDD 2014) %C Taiwan %D 2014 %F SisLab:1190 %P 400-412 %T A Novel Framework to Improve siRNA Efficacy Prediction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1190/ %V 8444 %X Short interfering RNA sequences (siRNAs) can knockdown target genes and thus have an immense impact on biology and pharmacy research. The key question of which siRNAs have high knockdown ability in siRNA research remains challenging as current known results are still far from expectation. This work aims to develop a generic framework to enhance siRNA knockdown efficacy prediction. The key idea is first to enrich siRNA sequences by incorporating them with rules found for designing effective siRNAs and representing them as transformed matrices, then to employ the bilinear tensor regression to do prediction on those matrices. Experiments show that the proposed method achieves results better than existing models in most cases. %0 Journal Article %@ 1471-2105 %A Bui, Ngoc Thang %A Ho, Tu Bao %A Kanda, T.A. %A VNU-UET, %D 2015 %F SisLab:1468 %J BMC Bioinformatics %N 80 %T Semi-supervised Tensor Regression Model for siRNA Efficacy Prediction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1468/ %V 16 %X Short interfering RNAs (siRNAs) can knockdown target genes and thus have an immense impact on biology and pharmacy research. The key question of which siRNAs have high knockdown ability in siRNA research remains challenging as current known results are still far from expectation. This work aims to develop a generic framework to enhance siRNA knockdown efficacy prediction. The key idea is first to enrich siRNA sequences by incorporating them with rules found for designing effective siRNAs and representing them as enriched matrices, then to employ the bilinear tensor regression to predict knockdown efficacy of those matrices. Experiments show that the proposed method achieves better results than existing models in most cases. Our model not only provides a suitable siRNA representation but also can predict siRNA efficacy more accurate and stable than most of state–of–the–art models. Source codes are freely available on the web at: http://​www.​jaist.​ac.​jp/​\~bao/​BiLTR/ %0 Conference Paper %A Bui, Ngoc Thang %A Le, Sy Vinh %A Ho, Tu Bao %B KSE: the 2015 International Conference on Knowledge and Systems Engineering %C Ho Chi Minh city, Vietnam %D 2015 %F SisLab:1632 %P 120-125 %T MVRM: A Hybrid Approach to Predict siRNA Efficacy %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1632/ %X The discovery of RNA interference (RNAi) leads to design novel drugs for different diseases. Selecting short interfering RNAs (siRNAs) that can knockdown target genes efficiently is one of the key tasks in studying RNAi. A number of predictive models have been proposed to predict knockdown efficacy of siRNAs, however, their performance is still far from the expectation. This work aims to develop a predictive model to enhance siRNA knockdown efficacy prediction. The key idea is to combine both the rule -- based and the model -- based approaches. To this end, views of siRNAs that integrate available siRNA design rules are first learned using an adaptive Fuzzy C Means (FCM) algorithm. The learned views and other properties of siRNAs are combined to final representations of siRNAs. The elastic net regression method is employed to learn a predictive model from these final representations. Experiments on benchmark datasets showed that the proposed method achieved stable and accurate results in comparison with other methods. %0 Conference Paper %A Bui, Phi Diep %A Nguyen, Trong Khanh %A Nguyen, Viet Ha %B SoICT: the 2013 Symposium on Information and Communication Technology %C Danang, Vietnam %D 2013 %F SisLab:1595 %P 261-266 %T A runtime approach for estimating resource usage %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1595/ %X In the era of information explosion, a program is necessary to be scalable. Therefore, scalability analysis becomes very important in software verification and validation. However, current approaches to empirical scalability analysis remain limitations related to the number of supported models and performance. In this paper, we propose a runtime approach for estimating the program resource usage with two aims: evaluating the program scalability and revealing potential errors. In this approach, the resource usage of a program is first observed when it is executed on inputs with different scales, the observed results are then fitted on a model of the usage according to the program's input. Comparing to other approaches, ours supports diverse models to illustrate the resource usage, i.e., linear-log, power-law, polynomial, etc. We currently focus on the computation cost and stack frames usage as two representatives of resource usage, but the approach can be extended to other kinds of resource. The experimental result shows that our approach achieves more precise estimation and better performance than other state-of-the-art approaches. %0 Report %9 Technical Report %A Bui, Quang Cuong %A Dinh, Tien Loc %A Luu, Anh Viet %A Nguyen, Viet Hoa %A Pham, Ngoc Quy %A Pham, Ngoc Hung %D 2018 %F SisLab:3202 %T A Toolchain for Source Code Quality Assurance of Java EE Applications %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3202/ %0 Conference Paper %A Bui, Quang Hung %A Nguyen, Thi Nhat Thanh %A Nguyen, Hai Chau %B The 2nd International Joint Conference (IJCC) %C Hanoi, Vietnam %D 2016 %F SisLab:2104 %P 125-130 %T Field Informatics: Application of ICT to environment management and monitoring %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2104/ %0 Journal Article %@ 0925-5001 %A Bui, Quoc Trung %A Vidal, Thibaut %A Ha, Minh Hoang %D 2019 %F SisLab:3509 %I Springer %J Journal of global optimization %N 1 %P 45-62 %T On three soft rectangle packing problems with guillotine constraints %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3509/ %V 74 %X We investigate how to partition a rectangular region of length L1 and height L2 into n rectangles of given areas (a1,…,an) using two-stage guillotine cuts, so as to minimize either (i) the sum of the perimeters, (ii) the largest perimeter, or (iii) the maximum aspect ratio of the rectangles. These problems play an important role in the ongoing Vietnamese land-allocation reform, as well as in the optimization of matrix multiplication algorithms. We show that the first problem can be solved to optimality in O(nlogn) , while the two others are NP-hard. We propose mixed integer linear programming formulations and a binary search-based approach for solving the NP-hard problems. Experimental analyses are conducted to compare the solution approaches in terms of computational efficiency and solution quality, for different objectives. %0 Report %9 Technical Report %@ TR2017-FIT-005 %A Bui, Thang %D 2017 %F SisLab:2916 %T Predictive models to predict knockdown efficacy of siRNAs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2916/ %0 Journal Article %A Bui, The Duy %A Nguyen, Thang Long %D 2007 %F SisLab:201 %J Sensors Journal, IEEE %K Accelerometers;Application software;Character recognition;Fingers;Handicapped aids;Mechanical sensors;Mechanical systems;Microelectronics;Micromechanical devices;Shape;accelerometers;biocommunications;fuzzy systems;gesture recognition;handicapped aids;knowledge based systems;microsensors;natural language processing;pattern classification;MEMS accelerometers;Vietnamese sign language;Vietnamese spelling rules;Vietnamese-based letters;finger signs;fuzzy rule-based classification system;microelectronic mechanical system accelerometers;posture recognition;Human computer interaction;Vietnamese sign language (VSL);microelectronic mechanical system (MEMS) sensors;sign language recognition; %N 5 %P 707-712 %T Recognizing Postures in Vietnamese Sign Language With MEMS Accelerometers %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/201/ %V 7 %X In this paper, we discuss the application of microelectronic mechanical system (MEMS) accelerometers for recognizing postures in Vietnamese Sign Language (VSL). We develop a similar device to the Accele Glove for the recognition of VSL. In addition to the five sensors as in the Accele Glove, we placed one more sensor on the back of the hand to improve the recognition process. In addition, we use a completely different method for the classification process leading to very promising results. This paper concentrates on signing with postures, in which the user spells each word with finger signs corresponding to the letters of the alphabet. Therefore, we focus on the recognition of postures that represent the 23 Vietnamese-based letters together with two postures for ``space'' and ``punctuation''. The data obtained from the sensing device is transformed to relative angles between fingers and the palm. Each character is recognized by a fuzzy rule-based classification system, which allows the concept of vagueness in recognition. In addition, a set of Vietnamese spelling rules has been applied to improve the classification results. The recognition rate is high even when the postures are not performed perfectly, e.g., the finger is not bended completely or the palm is not straight %0 Journal Article %@ 0925-2312 %A Bui, The Duy %A Nong, Thi-Hoa %A Dang, Trung Kien %D 2015 %F SisLab:1481 %I Elsevier %J Neurocomputing %P 59-64 %T Improving Learning Rule for Fuzzy Associative Memory with Combination of Content and Association %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1481/ %V 149 %0 Conference Paper %A Bui, Van Tan %A Nguyen, Phuong Thai %B 2018 10th International Conference on Knowledge and Systems Engineering (KSE) (KSE'18) %C Ho Chi Minh city, Vietnam %D 2018 %F SisLab:3407 %T Enhancing Performance of Lexical Entailment Recognition for Vietnamese based on Exploiting Lexical Structure Features %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3407/ %X The lexical entailment recognition problem aims to identify the is-a relation between words. The problem has recently been receiving research attention in the natural language processing field. In this study, we propose a novel method (VLER) for this problem on Vietnamese. For this purpose, we first exploit such lexical structure information of words as a feature, then combine this feature with vectors representation of words such as a unique feature for recognizing the relation. Moreover, we applied a number of methods based on word embedding and supervised learning, experimental results showed that our method achieves the best performance in the hypernymy detection task than other methods in terms of accuracy. %0 Conference Paper %A Bui, Van Tan %A Nguyen, Phuong Thai %A Dinh, Khac Quy %A University of Economic and Technical Industries, Hanoi, Vietnam, %A VNU University of Engineering and Technology, Hanoi, Vietnam, %A VNU University of Engineering and Technology, Hanoi, Vietnam, %B Fundamental and Applied IT Research Conference (Fair), 2018 %D 2018 %F SisLab:3412 %T ĐO ĐỘ TƯƠNG TỰ NGỮA NGHĨA CỦA CẶP NGÔN NGỮ ANH-VIỆT THEO MÔ HÌNH PHÂN PhỐI NGỮ NGHĨA SONG NGỮ %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3412/ %X Đo lường độ tương tự ngữ nghĩa giữa các từ là một bài toán nghiên cứu cốt lõi và có nhiều ứng dụng trong xử lý ngôn ngữ tự nhiên. Những nghiên cứu được công bố gần đây thường giải quyết bài toán này cho đơn ngữ. Gần đây, chứng kiến sự gia tăng không ngừng số lượng những ứng dụng xử lý tự nhiên đa ngôn ngữ, đặt ra yêu cầu cần có các kỹ thuật đo lường độ tương tự ngữ nghĩa song ngữ một cách hiệu quả. Trong bài viết này, chúng tôi trình bày một số kỹ thuật đo độ tương tự ngữ nghĩa song ngữ theo tiếp cận nhúng từ song ngữ; đề xuất một mô hình mạng nơron xây dựng không gian vector biểu diễn ngữ nghĩa song ngữ; xây dựng bộ dữ liệu chuẩn cho bài toán đo độ tương tự ngữ nghĩa song ngữ Việt-Anh. Cuối cùng, chúng tôi thực nghiệm và đánh giá các kỹ thuật trên bộ dữ liệu đã xây dựng. %0 Conference Paper %A Bui, Van Tan %A Nguyen, Phuong Thai %A Pham, Van Lam %A University of Economic and Technical Industries, Hanoi, Vietnam, %A VNU University of Engineering and Technology, Hanoi, Vietnam, %A Institute of Linguistics, Vietnam Academy of Social Sciences, Hanoi, Vietnam, %B International Conference on Computational Linguistics and Intelligent Text Processing %D 2018 %F SisLab:3411 %T Hypernymy Detection for Vietnamese Using Dynamic Weighting Neural Network %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3411/ %X The hypernymy detection problem aims to identify the "is-a" relation between words. The problem has recently been receiving attention from researchers in the field of natural language processing. So far, fairly-effective methods for hypernymy detection in English have been reported. Studies of hypernymy detection in Vietnamese have not been reported yet. In this study, we applied a number of hypernymy detection methods based on word embeddings and supervised learning for Vietnamese. We propose an improvement on the method given by Luu Tuan Anh et al. (2016) by weighting context words proportionally to the semantic similarity between them and the hypernym. Based on Vietnamese WordNet, three datasets for hypernymy detection were built. Experimental results showed that our proposal can increase the efficiency from 8% to 10% in terms of accuracy compared to the original method. %0 Conference Paper %A Bui Minh, Tuan %A Ta Duc, Tuyen %A Nguyen Linh, Trung %A Nguyen Viet, Ha %B IEEE Wireless Communications and Networking Conference (WCNC) %C Seoul, South Korea %D 2020 %F SisLab:4001 %T Autoencoder based friendly jamming %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4001/ %X Physical layer security (PLS) provides lightweight security solutions in which security is achieved based on the inherent random characteristics of the wireless medium. In this paper, we consider the PLS approach called friendly jamming (FJ), which is more practical thanks to its low computational complexity. State-of-the-art methods require that legitimate users have full channel state information(CSI)of their channel.Thanks to the recent promising application of the autoencoder (AE) in communication, we propose a new FJ method for PLS using AE without prior knowledge of the CSI. The proposed AE-based FJ method can provide good secrecy performance while avoiding explicit CSI estimation. We also apply the recently proposed tool for mutual information neural estimation (MINE) to evaluate the secrecy capacity. Moreover, we leverage MINE to avoid end-to-end learning in AE-based FJ. %0 Journal Article %@ 0377-2217 %A Bulhões, Teobaldo %A Ha, Minh Hoang %A Martinelli, Rafael %A Vidal, Thibaut %A Instituto de Computação, Universidade Federal Fluminense, Niterói, Brazil, %A University of Engineering and Technology, Vietnam National University, Hanoi, Viet Nam, %A Departamento de Engenharia Industrial, Pontifícia Universidade Católica do Rio de Janeiro, Brazil, %A Departamento de Informática, Pontifícia Universidade Católica do Rio de Janeiro, Brazil, %D 2017 %F SisLab:2574 %I Elsevier %J European Journal of Operational Research %T The vehicle routing problem with service level constraints %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2574/ %X We consider a vehicle routing problem which seeks to minimize cost subject to service level constraints on several groups of deliveries. This problem captures some essential challenges faced by a logistics provider which operates transportation services for a limited number of partners and should respect contractual obligations on service levels. The problem also generalizes several important classes of vehicle routing problems with profits. To solve it, we propose a compact mathematical formulation, a branch-and-price algorithm, and a hybrid genetic algorithm with population management, which relies on problem-tailored solution representation, crossover and local search operators, as well as an adaptive penalization mechanism establishing a good balance between service levels and costs. Our computational experiments show that the proposed heuristic returns very high-quality solutions for this difficult problem, matches all optimal solutions found for small and medium-scale benchmark instances, and improves upon existing algorithms for two important special cases: the vehicle routing problem with private fleet and common carrier, and the capacitated profitable tour problem. The branch-and-price algorithm also produces new optimal solutions for all three problems. %0 Report %9 Technical Report %A Can, Duy Cat %D 2020 %F SisLab:3983 %T Exploiting document graphs for inter-sentence relation extraction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3983/ %X Background: Most previous relation extraction (RE) studies have focused on intra-sentence relations and haveignored relations that span sentences, i.e. inter-sentence relations. Such relations connect entities at thedocument level rather than as relational facts in a single sentence. Extracting facts that are expressed acrosssentences leads to some challenges and requires different approaches than those usually applied in recentintra-sentence relation extraction. Despite recent results, there are still limitations to be overcome.Results:We present a novel representation for a sequence of consecutive sentences, namely documentsub-graph, to extract inter-sentence relations. Experiments on the BioCreative V Chemical-Disease Relationcorpus demonstrate the advantages and robustness of our novel system to extract both intra- andinter-sentence relations in biomedical literature abstracts. The experimental results are comparable tostate-of-the-art approaches and show the potential by demonstrating the effectiveness of graphs, deeplearning-based model and other processing techniques. Experiments were also carried out to verify therationality and impact of various additional information and model components.Conclusions:Our proposed graph-based representation helps to extract∼50% of inter-sentence relations andboosts the model performance on both precision and recall compared to the baseline model. %0 Conference Paper %A Can, Duy Cat %A Ho, Thi Nga %A Chng, Eng Siong %B International Conference on Asian Language Processing (IALP 2018 ) %C Bandung, Indonesia %D 2018 %F SisLab:3169 %T A Hybrid Deep Learning Architecture for Sentence Unit Detection %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3169/ %X Automatic speech recognition systems currently deliver an unpunctuated sequence of words which is hard to peruse for human and degrades the performance of the downstream natural language processing tasks. In this paper, we propose a hybrid approach for Sentence Unit Detection, in which the focus is on adding the full stop [.] to the unstructured text. Our model profits from the advantage of two dominant deep learning architectures: (i) the ability to learn the long dependencies in both directions of a bidirectional Long Short-Term Memory; (ii) the ability to capture the local context with Convolutional Neural Networks. We also empirically study the training objective of our networks using extra-loss and further investigate the impacts of each model component on the overall result. Experiments conducted on two large-scale datasets demonstrated that the proposed architecture outperforms previous separated methods by a substantial margin of 1.82-1.91% of F1. %0 Book Section %A Can, Duy Cat %A Le, Hoang Quynh %A Ha, Quang Thuy %A VNU-UET, %B ACIIDS 2021: Intelligent Information and Database Systems %D 2021 %F SisLab:4485 %I Springer %P 386-398 %T Detection of Distributed Denial of Service Attacks Using Automatic Feature Selection with Enhancement for Imbalance Dataset %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4485/ %X With the development of technology, the highly accessible internet service is the biggest demand for most people. Online network, however, has been suffering from malicious attempts to disrupt essential web technologies, resulting in service failures. In this work, we introduced a model to detect and classify Distributed Denial of Service attacks based on neural networks that take advantage of a proposed automatic feature selection component. The experimental results on CIC-DDoS 2019 dataset have demonstrated that our proposed model outperformed other machine learning-based model by large margin. We also investigated the effectiveness of weighted loss and hinge loss on handling the class imbalance problem. %0 Conference Paper %9 Technical Report %A Can, Duy Cat %A Le, Hoang Quynh %A Ha, Quang Thuy %B ACIIDS-2021 %D 2021 %F SisLab:4289 %I ACIIDS 2021 %T Detection of Distributed Denial of Service Attacks using Automatic Feature Selection with Enhancement for Imbalance Dataset %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4289/ %X Abstract: With the development of technology, the highly accessible internet service is the biggest demand for most people. Online network, however, has been suffering from malicious attempts to disrupt essential web technologies, resulting in service failures. In this work, we introduced a model to detect and classify Distributed Denial of Service attacks based on neural networks that take advantage of a proposed automatic feature selection component. The experimental results on CIC-DDoS 2019dataset have demonstrated that our proposed model outperformed other machine learning-based model by large margin. We also investigated the effectiveness of weighted loss and hinge loss on handling the class imbalance problem. %0 Conference Paper %A Can, Duy Cat %A Le, Hoang Quynh %A Ha, Quang Thuy %B 2019 11th Asian Conference on Intelligent Information and Database Systems (ACIIDS) %C Yogyakarta, Indonesia %D 2019 %F SisLab:3441 %P 140-152 %T Improving Semantic Relation Extraction System with Compositional Dependency Unit on Enriched Shortest Dependency Path %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3441/ %X Experimental performance on the task of relation extraction/classification has generally improved using deep neural network architectures. In which, data representation has been proven to be one of the most influential factors to the model's performance but still has many limitations. In this work, we take advantage of compressed information in the shortest dependency path (SDP) between two corresponding entities to classify the relation between them. We propose (i) a compositional embedding that combines several dominant linguistic as well as architectural features and (ii) dependency tree normalization techniques for generating rich representations for both words and dependency relations in the SDP. We also present a Convolutional Neural Network (CNN) model to process the proposed SDP enriched representation. Experimental results for both general and biomedical data demonstrate the effectiveness of compositional embedding, dependency tree normalization technique as well as the suitability of the CNN model. %0 Journal Article %A Cao, Thinh %A Yamada, Koichi %A Unehara, Muneyuki %A Suzuki, Izumi %A Nguyen, Do Van %D 2018 %F SisLab:3296 %J Computers %T Parallel Computation of Rough Set Approximations in Information Systems with Missing Decision Data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3296/ %X The paper discusses the use of parallel computation to obtain rough set approximations from large-scale information systems where missing data exist in both condition and decision attributes. To date, many studies have focused on missing condition data, but very few have accounted for missing decision data, especially in enlarging datasets. One of the approaches for dealing with missing data in condition attributes is named twofold rough approximations. The paper aims to extend the approach to deal with missing data in the decision attribute. In addition, computing twofold rough approximations is very intensive, thus the approach is not suitable when input datasets are large. We propose parallel algorithms to compute twofold rough approximations in large-scale datasets. Our method is based on MapReduce, a distributed programming model for processing large-scale data. We introduce the original sequential algorithm first and then the parallel version is introduced. Comparison between the two approaches through experiments shows that our proposed parallel algorithms are suitable for and perform efficiently on large-scale datasets that have missing data in condition and decision attributes. %0 Journal Article %@ 2296-7745 %A Christopher, D. Elvidge %A Tilottama, Ghosh %A Kimberly, Baugh %A Mikhail, Zhizhin %A Feng-Chi, Hsu %A Nilo, Selim Katada %A Wilmon, Penalosa %A Bui, Quang Hung %D 2018 %F SisLab:3358 %J Frontiers in Marine Science %T Rating the Effectiveness of Fishery Closures With Visible Infrared Imaging Radiometer Suite Boat Detection Data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3358/ %X Fishery closures are widely used to promote the sustainability of fish stocks. Fishery agencies typically have very little data relevant to planning closure enforcement actions and evaluating the effectiveness of closures, due in part to the vast expanse and remote nature of many closures. In some cases the effectiveness of closures can be evaluated using data from GPS based beacons, such as Automatic Identification System (AIS) or Vessel Monitoring Systems (VMS) installed on fishing boats. In fisheries where few boats are equipped with AIS or VMS, the rating of closures relies on other data sources capable of detecting or inferring fishing activity. One such source comes from low light imaging data collected by the NASA/NOAA Visible Infrared Imaging Radiometer Suite (VIIRS), which can detect fishing boats using lights to attract catch. This is a widely used practice in Asia and several other regions. NOAA has developed an automatic system for reporting the locations of VIIRS boat detections with a nominal 4 h temporal latency. VIIRS boat detection alerts are running for more than 900 fishery closures in the Philippines, with email and SMS transmission modes. These alerts are being actively used in the Philippines to plan enforcement actions and there is a growing list of apprehensions that occurred based on tip-offs from VIIRS. The VIIRS boat detection archive extends back to April 2012. A VIIRS closure index (VCI) has been developed to rate the effectiveness of closures on monthly increments in terms of a percentage. The VCI analysis was performed on three types of closures: an ad hoc fishery closure associated with a toxic industrial discharge, a seasonal fishery closure and a permanent closure in restricted coastal waters. The VCI results indicate that it is possible to rank the effectiveness of different closure, year-to-year differences in compliance levels, and to identify closure encroachments which may warrant additional enforcement effort. %0 Conference Paper %A Chu, Minh Hue %A Dang, Duc Hanh %B Proc. 12nd Int. Conf. Knowledge and Systems Engineering (KSE) %D 2020 %F SisLab:4517 %P 109-114 %T Automatic Extraction of Analysis Classes from Use Cases %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4517/ %0 Journal Article %@ 1859-1043 %A Chu, Minh Hue %A Dang, Duc Hanh %D 2011 %F SisLab:1973 %J Tạp chí Nghiên cứu Khoa học & Công nghệ Quân sự %N 16 %P 163-168 %T A Domain Specific Language for the RBAC Domain %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1973/ %V 12 %0 Conference Paper %A Chu, Minh Hue %A Dang, Duc Hanh %A Nguyen, Ngoc Binh %B 8th National Conference on Fundamental and Applied Information Technology Research (FAIR) %C Hanoi, Vietnam %D 2015 %F SisLab:1971 %P 590-599 %T A Method to Generate Test Cases from Use Cases %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1971/ %0 Conference Paper %A Chu, Minh Hue %A Dang, Duc Hanh %A Nguyen, Ngoc Binh %B The 9th National Conference on Fundamental and Applied Information Technology Research (FAIR) %C Can Tho, Vietnam %D 2016 %F SisLab:1970 %P 1-8 %T A Method to Specify Software Functional Requirements for System Test Case Generation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1970/ %0 Conference Paper %A Chu, Minh Hue %A Dang, Duc Hanh %A Nguyen, Ngoc Binh %B KSE: the 2018 International Conference on Knowledge and Systems Engineering %C Vietnam %D 2018 %F SisLab:3142 %P 252-257 %T A Transformation-Based Method for Test Case Automatic Generation from Use Cases %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3142/ %0 Journal Article %@ 0350-5596 %A Chu, Minh Hue %A Dang, Duc Hanh %A Nguyen, Ngoc Binh %A Le, Minh Duc %D 2018 %F SisLab:3139 %I Slovensko Drustvo Informatika %J Informatica - Journal of Computing and Informatics %N 3 %P 325-343 %T USL: A Domain-Specific Language for Precise Specification of Use Cases and Its Transformations %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3139/ %V 42 %X A use case model is often represented by a UML use case diagram and loosely structured textual descriptions. The use case model expressed in such a form contains ambiguous and imprecise parts. This prevents integrating it into model-driven approaches, where use case models are often taken as the source of transformations. In this paper, we introduce a domain-specific language named the Use case Specification Language (USL) to precisely specify use cases. We define the abstract syntax of USL using a metamodel together with OCL wellformedness rules and then provide a graphical concrete syntax for the usability goal. We also define a precise semantics for USL by mapping USL models to Labelled Transition Systems (LTSs). It opens a possibility to transform USL models to software artifacts such as test cases and design models. We focus on a transformation from a USL model to a template-based use case description in order to illustrate our method. A language evaluation of USL is also performed in this paper. %0 Conference Paper %A Chu, Minh Hue %A Dang, Duc Hanh %A Nguyen, Ngoc Binh %A Le, Minh Duc %A Nguyen, Thi Hanh %B 8th International Symposium on Information and Communications (SoICT) %C Nha Trang %D 2017 %F SisLab:2673 %P 1-8 %T USL: Towards Precise Specification of Use Cases for Model-Driven Development %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2673/ %0 Journal Article %@ 0218-1940 %A Chu, Minh Hue %A Dang, Duc Hanh %A Nguyen, Ngoc Binh %A Truong, Anh Hoang %D 2019 %F SisLab:3580 %I World Scientific Publishing %J International journal of software engineering and knowledge engineering (IJSEKE) %N 9 %P 1313-1345 %T USLTG: Test Case Automatic Generation by Transforming Use Cases %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3580/ %V 29 %X This paper proposes a transformation-based method to automatically generate functional test cases from use cases named USLTG (USL-based Test Generation). We first focus on developing a modeling language named Test Case Specification Language (TCSL) in order to express test cases. Test cases in TCSL can contain detailed information including test steps, test objects within steps, actions of test objects, and test data. Such information is often ignored in currently available test case specifications. We then aim to generate test cases in a TCSL model by a transformation from use cases that are represented by a Use case Specification Language (USL). The USLTG transformation includes three main steps, in generating: (1) scenarios, (2) test data, and (3) a TCSL model. Within our transformation, the OCL solver is employed in order to build system snapshots as the part of test cases and to identify other test data. We applied our method to two case studies and evaluated our method by comparing with other recent works. %0 Conference Paper %A Chu, Thi Minh Hue %A Dang, Duc Hanh %A Nguyen, Ngoc Binh %B Hội nghị khoa học quốc gia lần thứ VIII “Nghiên cứu cơ bản và ứng dụng Công nghệ Thông tin” năm 2015 (FAIR 2015) %C Hanoi, Vietnam %D 2015 %F SisLab:1737 %T Phương pháp sinh tự động ca kiểm thử từ mô hình ca sử dụng(Automatic Generation of Test Cases based on Use Cases Model) %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1737/ %0 Conference Paper %A Chu, Thi Minh Hue %A Nguyen, Ngoc Binh %A Dang, Duc Hanh %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1515 %T Kiểm thử dựa trên mô hình với cách tiếp cận mô hình hóa miền chuyên biệt %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1515/ %X Trong phát triển phần mềm, kiểm thử là một quy trình quan trong và không thể thiếu để phát triển một hệ thống phần mềm tin cậy và có chất lượng cao. Để có thể giảm chi phí, tăng chất lượng cho hoạt động kiểm thử một hướng tiếp cận cho kiểm thử phần mềm là kỹ thuật kiểm thử dựa trên mô hình (Model-Based Tesing - MBT). Kỹ thuật này cho phép sinh tự động các ca kiểm thử và thực thi tự động các ca kiểm thử từ mô hình. Trong nghiên cứu này, chúng tôi tập trung vào nghiên cứu quy trình sinh tự động (Automated Test Cases Generation) các ca kiểm thử từ mô hình với hướng tiếp cận mô hình hóa miền chuyên biệt (Domain-Specific Modeling - DSM) cho vấn đề sinh các ca kiểm thử từ đặc tả ca sử dụng (Use Case Specification). Đặc tả ca sử dụng được viết trong tài liệu đặc tả phần mềm bằng ngôn ngữ tự nhiên vì vậy mà việc sinh tự động các ca kiểm thử từ đặc tả này vẫn còn nhiều thách thức. Chất lượng sinh các ca kiểm thử phụ thuộc vào tiêu chí phủ các ca kiểm thử. Nghiên cứu này của chúng tôi tập chung vào nghiên cứu và đề xuất phương pháp sinh tự động các ca kiểm thử từ đặc tả ca sử dụng với độ phủ các ca kiểm thử đạt được các tiêu chí phủ kiểm thử %0 Conference Paper %A Chung, Jae Du %A Lee, Yang Koo %A Vu, Thi Hong Nhan %B Korea Spatial Information System Society %D 2006 %F SisLab:2140 %T A Spatiotemporal Data Management and Mining System Architecture %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2140/ %0 Conference Paper %A Collier, Nigel %A Paster, Ferdinand %A Tran, Mai Vu %B 2014: the 5th International Workshop on Health Text Mining and Information Analysis (EACL) %C Gothenburg, Sweden %D 2014 %F SisLab:1308 %P 11-20 %T The Impact of Near Domain Transfer on Biomedical Named Entity Recognition %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1308/ %0 Journal Article %@ 1932-6203 %A Collier, Nigel %A Tran, Mai Vu %A Le, Hoang Quynh %A Ha, Quang Thuy %A Oellrich, Anika %A Rebholz-Schuhmann, Dietrich %D 2013 %F SisLab:188 %J PLOS ONE %T Mining Phenotype Candidates from the Scientific Literature with Diverse Hypothesis Resolution Strategies %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/188/ %0 Conference Paper %A Cuong, Le Ba %A Pham, Ngoc Hung %A VNU-UET, %B 2012 Fourth International Conference on Knowledge and Systems Engineering (KSE) %C Danang, Vietnam %D 2012 %F SisLab:102 %K Adaptation models;Automata;Doped fiber amplifiers;Modeling;Software;Unified modeling language;deterministic automata;finite automata;formal verification;object-oriented programming;software quality;Thompson algorithm;black-box components;component alphabet;component-based software quality;component-based systems;formal models;minimal deterministic finite automaton;model checking;model generation;model-based approaches;model-based testing;Thompson algorithm;black-box components;model generation;model-based approaches; %P 217 -222 %T A Method for Generating Models of Black-Box Components %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/102/ %X The model-based approaches are difficult to be applied in practice due to the lack of formal models describing behaviors of systems. This paper proposes a method for generating accurate models of components in order to solve this issue in the context of the component-based systems. The key idea of this method is to generate minimal deterministic finite automata as the accurate models of the corresponding components. For this purpose, the proposed method first computes a set of traces as a regular language of a given component by executing all possible experiments over the alphabet of the component. This method then constructs a regular expression to represent this set. After that, a minimal deterministic finite automaton as an accurate model of the component is generated by applying the Thompson algorithm and some optimized activities. The generated models are useful for the existing model-based approaches, e.g., model checking and model-based testing in improving quality of component-based software. An implemented tool supporting the method and experimental results are also presented. %0 Conference Paper %A Dam, Dinh Hiep %A Le, Hoai Nam %A Bui, Duy Toan %A Nguyen, Ngoc Linh %B International Conference on Engineering Mechanics and Automation (ICEMA) %C Hanoi, Vietnam %D 2019 %F SisLab:3656 %T A Research on Conveyor Belt 3D Printer in Industrial Applications %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3656/ %X A 3d printing technique is an additive manufacturing technique where 3D objects and parts are made by the addition of multiple layers of material.[1] It is a type of rapid prototyping. The layers are stacked up in a variety of ways depending on the technology being used. It can use a wide range of materials such as ABS, PLA, and composites as well. This technology allows the design of complex components, therefore, avoiding assembly requirements at no additional cost. This paper presents research on 3D printer design with a conveyor belt, which can be applied in industrial applications. Firstly, a CAD model of a 3d printer has been created using SolidWorks. In this step, all the parts of the model are designed and then are assembled in the SolidWorks workbench to create the 3D printer assembly. Then, the electronics board which controls the entire printing process is studied and integrated into the mechanical structure. The electronics board compiles the STL file to a suitable form to carry out the printing process and is connected to the PC using a USB-to-serial converter. Finally, a portable 3D printer is built and tested with various types of models. %0 Journal Article %A Dang, Cao Cuong %A Le, Si Quang %A Gascuel, Olivier %A Le, Sy Vinh %D 2010 %F SisLab:2578 %J BMC Evolutionary Biology %N 1 %P 99 %T FLU, an Amino Acid Substitution Model for Influenza Proteins %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2578/ %V 10 %X The amino acid substitution model is the core component of many protein analysis systems such as sequence similarity search, sequence alignment, and phylogenetic inference. Although several general amino acid substitution models have been estimated from large and diverse protein databases, they remain inappropriate for analyzing specific species, e.g., viruses. Emerging epidemics of influenza viruses raise the need for comprehensive studies of these dangerous viruses. We propose an influenza-specific amino acid substitution model to enhance the understanding of the evolution of influenza viruses. %0 Journal Article %@ 1471-2105 %A Dang, Cao Cuong %A Le, Sy Vinh %A Gascuel, Oliver %A Hazes, Bart %A Le, Si Quang %D 2014 %F SisLab:1298 %J BMC Bioinformatics %T FastMG: a simple, fast, and accurate maximum likelihood procedure to estimate amino acid replacement rate matrices from large data sets %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1298/ %X Amino acid replacement rate matrices are a crucial component of many protein analysis systems such as sequence similarity search, sequence alignment, and phylogenetic inference. Ideally, the rate matrix reflects the mutational behavior of the actual data under study; however, estimating amino acid replacement rate matrices requires large protein alignments and is computationally expensive and complex. As a compromise, sub-optimal pre-calculated generic matrices are typically used for protein-based phylogeny. Sequence availability has now grown to a point where problem-specific rate matrices can often be calculated if the computational cost can be controlled. %0 Conference Paper %A Dang, Cao Cuong %A Le, Sy Vinh %A Le, Si Quang %B International Conference on Knowledge and Systems Engineering, 2009. KSE '09 %D 2009 %F SisLab:67 %K biology computing, influenza-specific amino acid substitution model, maximum likelihood approach, maximum likelihood estimation, phylogenetic trees, protein databases, protein phylogenetic analysis, protein sequence alignment, protein sequence comparative systems, protein sequence similarity searching, Proteins, tree topologies, trees (mathematics) %P 19 -25 %T Influenza-specific Amino Acid Substitution Model %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/67/ %X Amino acid substitution model is a crucial component in protein sequence comparative systems such as protein sequence similarity searching, protein sequence alignment, and protein phylogenetic analysis. Although several general amino acid substitution models have been estimated from large protein databases, they might not be appropriate for analyzing specific species. In this paper, we apply the maximum likelihood approach to all influenza protein sequences to estimate an amino acid substitution model of so-called I09 for influenza viruses. Comparing I09 with fourteen other widely used models, we achieve remarkable results: (1) a likelihood improvement of phylogenetic trees based on I09 compared with other models. Precisely, I09 results in the best likelihood in 436 out of 489 cases tested; (2) tree topologies constructed with I09 and other models are frequently different indicating that the impact of I09 is not only on the likelihood improvement but also in tree topologies; (3) marked differences between I09 and other models revealing that existing models are not be able to capture the amino acid substitution process of influenza viruses. %0 Journal Article %A Dang, Cao Cuong %A Lefort, Vincent %A Le, Sy Vinh %A Le, Si Quang %A Gascuel, Olivier %D 2011 %F SisLab:72 %J Bioinformatics %N 19 %P 2758-2760 %T ReplacementMatrix: a Web Server for Maximum-Likelihood Estimation of Amino Acid Replacement Rate Matrices %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/72/ %V 27 %X Amino acid replacement rate matrices are an essential basis of protein studies (e.g. in phylogenetics and alignment). A number of general purpose matrices have been proposed (e.g. JTT, WAG, LG) since the seminal work of Margaret Dayhoff and co-workers. However, it has been shown that matrices specific to certain protein groups (e.g. mitochondrial) or life domains (e.g. viruses) differ significantly from general average matrices, and thus perform better when applied to the data to which they are dedicated. This Web server implements the maximum-likelihood estimation procedure that was used to estimate LG, and provides a number of tools and facilities. Users upload a set of multiple protein alignments from their domain of interest and receive the resulting matrix by email, along with statistics and comparisons with other matrices. A non-parametric bootstrap is performed optionally to assess the variability of replacement rate estimates. Maximum-likelihood trees, inferred using the estimated rate matrix, are also computed optionally for each input alignment. Finely tuned procedures and up-to-date ML software (PhyML 3.0, XRATE) are combined to perform all these heavy calculations on our clusters. %0 Journal Article %A Dang, Cao Cuong %A Peón, Antonio %A Ballester, Pedro %D 2018 %F SisLab:2958 %J BMC Medical Genomics %N 1 %P 10 %T Unearthing new genomic markers of drug response by improved measurement of discriminative power %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2958/ %V 11 %X Oncology drugs are only effective in a small proportion of cancer patients. Our current ability to identify these responsive patients before treatment is still poor in most cases. Thus, there is a pressing need to discover response markers for marketed and research oncology drugs. Screening these drugs against a large panel of cancer cell lines has led to the discovery of new genomic markers of in vitro drug response. However, while the identification of such markers among thousands of candidate drug-gene associations in the data is error-prone, an appraisal of the effectiveness of such detection task is currently lacking. %0 Thesis %9 PhD %A Dang, Duc Hanh %B Department of Computer Science %D 2009 %F SisLab:1961 %I University of Bremen %T On Integrating Triple Graph Grammars and OCL for Model-Driven Development %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1961/ %0 Conference Paper %A Dang, Duc Hanh %B The 4th International Conference on Graph Transformations (ICGT) %D 2008 %F SisLab:1964 %I Springer %P 481-483 %T Triple Graph Grammars and OCL for Validating System Behavior %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1964/ %V 5214 %0 Conference Paper %A Dang, Duc Hanh %B The 10th International Conference on Model-Driven Engineering Languages and Systems (MoDELS/Doctoral Symposium) %D 2007 %F SisLab:1968 %I CEUR-WS.org %P 1-5 %T Validation of System Behavior from an Integrated Semantics of Use Case and Design Models %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1968/ %V 262 %0 Conference Paper %A Dang, Duc Hanh %A Cabot, Jordi %B 2013: the 20th Asia-Pacific Software Engineering Conference (APSEC 2013) %C Bangkok, Thailand %D 2013 %F SisLab:771 %T Automating Inference of OCL Business Rules from User Scenarios %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/771/ %0 Conference Paper %A Dang, Duc Hanh %A Cabot, Jordi %B 2014: The 6th International Conference on Knowledge and Systems Engineering (KSE 2014) %C Hanoi, Vietnam %D 2014 %F SisLab:441 %P 219-231 %T On Automating Inference of OCL Constraints from Counterexamples and Examples %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/441/ %0 Conference Paper %A Dang, Duc Hanh %A Gogolla, M. %A VNU-UET, %B 2012 Fourth International Conference on Knowledge and Systems Engineering (KSE) %C Danang, Vietnam %D 2012 %F SisLab:103 %K Automata;Computational modeling;Grammar;Quality assurance;Strontium;Unified modeling language;formal specification;graph grammars;program verification;software quality;OCL-based framework;model transformation;model-driven approach;object constraint language;on-the-fly verification;specification method;transformation quality assurance;transformation validation;transformation verification;triple graph grammar;Graph Transformation;Invariant;Model Transformation;OCL;Pre- and Postcondition;Validation and Verification; %P 223 -230 %T An Approach for Quality Assurance of Model Transformations %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/103/ %X Model transformation is an important building block for model-driven approaches. It puts forward a necessity as well as a challenge for validating and verifying transformations. This paper proposes a specification method and an OCL-based framework for model transformations. The approach is based on an integration of Triple Graph Grammars and the Object Constraint Language (OCL) as a formal foundation. The OCL-based transformation framework offers an on-the-fly verification of model transformations and means for transformation quality assurance. %0 Journal Article %@ 0866-8612 %A Dang, Duc Hanh %A Gogolla, Martin %D 2016 %F SisLab:1952 %I VNU Journal of Science %J VNU Journal of Computer Science and Communication Engineering %N 1 %P 42-57 %T An OCL-based Framework for Model Transformations %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1952/ %V 32 %0 Conference Paper %A Dang, Duc Hanh %A Gogolla, Martin %B International Conference on Models in Software Engineering (MODELS) %D 2008 %F SisLab:1965 %I Springer %P 124-137 %T On Integrating OCL and Triple Graph Grammars %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1965/ %V 5421 %0 Conference Paper %A Dang, Duc Hanh %A Gogolla, Martin %B The 7th International Conference on Software Engineering and Formal Methods (SEFM) %D 2009 %F SisLab:1962 %I IEEE Computer Society %P 307-316 %T Precise Model-Driven Transformations Based on Graphs and Metamodels %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1962/ %0 Conference Paper %A Dang, Duc Hanh %A Truong, Anh Hoang %A Dang, Van Hung %B Proc. 5th Int. Conf. on New Trends in Software Methodologies, Tools and Technique (SoMeT) %D 2016 %F SisLab:1955 %I IOS Press %P 279-290 %T On Model Finding with Constraint Patterns %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1955/ %V 286 %0 Journal Article %@ 0948-695x %A Dang, Duc Hanh %A Truong, Anh Hoang %A Gogolla, Martin %D 2010 %F SisLab:1959 %J Journal of Universal Computer Science %N 17 %P 2293-2312 %T Checking the Conformance between Models Based on Scenario Synchronization %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1959/ %V 16 %0 Conference Paper %A Dang, Duc Hanh %A Truong, Anh Hoang %A Gogolla, Martin %B 8th International Conference on Automated Technology for Verification and Analysis (ATVA) %D 2010 %F SisLab:1960 %I Springer %P 97-111 %T On Scenario Synchronization %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1960/ %V 6252 %0 Journal Article %A Dang, Hai %A Nguyen, Thanh %A Pham, Minh Trang %A Dang, Cuong %A Hoang, Phuc %A Pham, Son %A Le, Vinh %A Le, Quang %A Phan, Hang %A Do, Dong %A Nguyen, Duc %D 2014 %F SisLab:2531 %J VNU Journal of Science: Computer Science and Communication Engineering %N 3 %T Preliminary results on the whole genome analysis of a Vietnamese individual %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2531/ %V 30 %X We present some preliminary results on the whole genome analysis of an anonymous Kinh Vietnamese (KHV) individual that was deeply sequenced to 34-fold using the Illumina sequencing technology. The sequenced genome covered 99.85 of the human reference genome (GRCh37). We discovered (1) 3408825 single polymorphism nucleotides (SNPs) of which 41396 (1.2) were novel, (2) 654024 short indels of which 35263 (5.4) were novel, i.e. not present in the dbSNP and 1000 genomes project databases. We also detected 10611 large SVs (the length ≥ 100bp) of four types, i.e. large indels (90.6), the inter-chromosomal translocations (3.1), the inversions (3.4) and the intra-chromosomal translocations (2.9). This study is our initial step toward large-scale projects on Vietnamese population. %0 Conference Paper %A Dang, Hai Ninh %A Nguyen, Quang Tan %A Nguyen, Ngoc Linh %B International Conference on Engineering Mechanics and Automation (ICEMA5) %C Hanoi, Vietnam %D 2019 %F SisLab:3655 %T A Research on Sensorless Control of Brushless DC Motor using Inductance Variation Technique %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3655/ %X Today, brushless DC motors (BLDCM) have been used in many applications replacing to brushed DC motors. Compared to brushed DC motors, BLDCM offer improved reliability, longer life, smaller size, and lower weight. Besides, BLDCM have become more popular in applications where efficiency is a critical concern and, generally speaking, a BLDCM is considered to be a high-performance motor capable of providing large amounts of torque over a wide speed range. In this paper, a research on sensorless control method which can drive a BLDCM smoothly from standstill to high speeds without position or speed sensors is carried out. Initial rotor position as well as speed of motor at a low speed range is estimated based on the inductance variation principle while at higher speed, the back EMF technique is applied. This sensorless control algorithm is modeled and simulated with MATLAB/SIMULINK software to verify the abilities of the method. The drive control scheme has been implemented on a single-chip controller (STM32F103) and experimental results reveal that the control procedure can work smoothly. %0 Journal Article %@ 2169-3536 %A Dang, Nam Khanh %A Ahmed, Akram Ben %A Abdallah, Abderazek Ben %A Tran, Xuan Tu %D 2020 %F SisLab:4049 %I IEEE %J IEEE Access %P 166642-166657 %T A thermal-aware on-line fault tolerance method for TSV lifetime reliability in 3D-NoC systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4049/ %V 8 %X Through-Silicon-Via (TSV) based 3D Integrated Circuits (3D-IC) are one of the most advanced architectures by providing low power consumption, shorter wire length and smaller footprint. However, 3D-ICs confront lifetime reliability due to high operating temperature and interconnect reliability, especially the Through-Silicon-Via (TSV), which can significantly affect the accuracy of the applications. In this paper, we present an online method that supports the detection and correction of lifetime TSV failures, named IaSiG. By reusing the conventional recovery method and analyzing the output syndromes, IaSiG can determine and correct the defective TSVs. Results show that within a group, $R$ redundant TSVs can fully localize and correct $R$ defects and support the detection of $R+1$ defects. Moreover, by using $G$ groups, it can localize up to GxR and detect up to Gx(R+1) defects. An implementation of IaSiG for 32-bit data in eight groups and two redundancies has a worst-case execution time (WCET) of 5,152 cycles while supporting at most 16 defective TSVs (50\% localization). By integrating IaSiG onto a 3D Network-on-Chip, we also perform a grid-search based empirical method to insert suitable numbers of redundancies into TSV groups. The empirical method takes the operating temperature as the factor of accelerated fault due to the fact that temperature is one of the major issues of 3D-ICs. The results show that the proposed method can reduce the number of redundancies from the uniform method while still maintaining the required Mean Time to Failure. %0 Journal Article %@ 2168-6750 %A Dang, Nam Khanh %A Ahmed, Akram Ben %A Okuyama, Yuichi %A Abdallah, Abderazek Ben %D 2017 %F SisLab:3940 %I IEEE %J IEEE Transactions on Emerging Topics in Computing %T Scalable design methodology and online algorithm for TSV-cluster defects recovery in highly reliable 3D-NoC systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3940/ %X 3D-Network-on-Chips exploit the benefits of Network-on-Chips and 3D-Integrated Circuits allowing them to be considered as one of the most advanced and auspicious communication methodologies. On the other hand, the reliability of 3D-NoCs, due to the vulnerability of Through Silicon Vias, still remains as a major problem. Most of the existing techniques rely on correcting the TSV defects by using redundancies or employing routing algorithms. Nevertheless, they are not suitable for TSV-cluster defects as they can either lead to costly area and power consumption overheads or they may result in non-minimal routing paths; thus, posing serious threats to the system reliability and overall performance. In this work, we present a scalable and low-overhead TSV usage and design method for 3D-NoC systems where the TSVs of a router can be utilized by its neighbors to deal with the cluster open defects. An adaptive online algorithm is also introduced to assist the proposed system to immediately work around the newly detected defects without using redundancies. The experimental results show the proposal ensure less than 2% of the routers being disabled, even with 50% of the TSV clusters defects. The performance evaluations also demonstrate unchanged performances for real applications under 5% of cluster defects. %0 Conference Paper %A Dang, Nam Khanh %A Ahmed, Akram Ben %A Tran, Xuan Tu %B 2019 IEEE 13th International Symposium on embedded Multicore/Manycore Systems-on-Chip (IEEE MCSoC-2019) %C Singapore %D 2019 %F SisLab:3557 %T An on-communication multiple-TSV defects detection and localization for real-time 3D-ICs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3557/ %X This paper presents “On Communication Through- Silicon-Via Test” (OCTT), an ECC-based method to localize faults without halting the operation of TSV-based 3D-IC systems. OCTT consists of two major parts named Statistical Detector and Isolation and Check. While Statistical Detector could detect open and short defects in TSVs that work without interrupting data transactions, the Isolation and Check algorithm enhances the ability to localize fault position. The Monte-Carlo simulations of Statistical Detector show x2 increment in the number of detected faults when compared to conventional ECC-based techniques. While Isolation and Ch eck helps localize the number of defects up to x4 and x5 higher. In addition, the worst case execution time is below 65,000 cycles with no performance degradation for testing which could be easily integrated into real-time applications. %0 Journal Article %A Dang, Nam Khanh %A Meyer, Michael %A Ahmed, Akram Ben %A Abdallah, Abderazek Ben %A Tran, Xuan Tu %D 2020 %F SisLab:3947 %J REV Journal on Electronics and Communications %N 1-2 %P 11-21 %T 2D Parity Product Code for TSV Online Fault Correction and Detection %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3947/ %V 10 %X Through-Silicon-Via (TSV) is one of the most promising technologies to realize 3D Integrated Circuits (3D-ICs).However, the reliability issues due to the low yield rates and the sensitivity to thermal hotspots and stress issues arepreventing TSV-based 3D-ICs from being widely and efficiently used. To enhance the reliability of TSV connections, usingerror correction code to detect and correct faults automatically has been demonstrated as a viable solution. This paperpresents a 2D Parity Product Code (2D-PPC) for TSV fault-tolerance with the ability to correct one fault and detect, at least,two faults. In an implementation of 64-bit data and 81-bit code-word, 2D-PPC can detect over 71 faults, on average. Itsencoder and decoder decrease the overall latency by 38.33% when compared to the Single Error Correction Double ErrorDetection code. In addition to the high detection rates, the encoder can detect 100% of its gate failures, and the decodercan detect and correct around 40% of its individual gate failures. The squared 2D-PPC could be extended using orthogonalLatin square to support extra bit correction %0 Journal Article %@ 2169-3536 %A Dang, Nam Khanh %A Meyer, Michael %A Ahmed, Akram Ben %A Abdallah, Abderazek Ben %A Tran, Xuan Tu %D 2020 %F SisLab:3942 %J IEEE Access %P 59571-59589 %T A non-blocking non-degrading multiple defects link testing method for 3D-Networks-on-Chip %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3942/ %V 8 %X As one of the most promising technologies to realize 3D Integrated Circuits (3D-ICs), Through-Silicon-Via (TSV) acts as the inter-layer link inside 3D Networks-on-Chip. However, the reliability issues due to the low yield rates and the sensitivity to thermal hotspots and stress issues are preventing TSV-based 3D-ICs from being widely and efficiently used. To ensure the correctness of TSV connections at run-time, detecting multiple (clustering) defects is an important feature. While Error Correction Codes are limited by a certain number of detectable faults, using Built-In-Self-Test (BIST) prevents the system from operating normally during the test time. This paper first presents a Parity Product Code (PPC) with the ability to correct one fault and detect, at least, two faults. Second, we present extended PPC (EPPC) to detect multiple defects within the links of Networks-on-Chip by using two or more additional matrices. Furthermore, we present the distance-aware version of EPPC to detect multiple defects by using only one extra matrix. The results show that the distance-aware EPPC can detect 100% of clustering defects and multiple random defects within two and three cycles, respectively. The performance evaluation for Network-on-Chip testing also shows no degradation while providing an extremely short response time (2-3 cycles). %0 Report %9 Technical Report %A Dang, Thanh Hai %D 2019 %F SisLab:3668 %T SKIPHOS: non-kinase specific phosphorylation site prediction with random forests and amino acid skip-gram embeddings %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3668/ %0 Conference Paper %A Dang, Thanh Hai %A Engelen, Kristof %A Meysman, Pieter %A Marchal, Kathleen %A Verschoren, Alain %A Laukens, Kris %B The 2009 International Conference on Knowledge and Systems Engineering (KSE) %D 2009 %F SisLab:2077 %P 7-12 %T Conditional Random Fields feature subset selection based on Genetic Algorithms for phosphorylation site prediction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2077/ %0 Journal Article %@ 1460-2059 %A Dang, Thanh Hai %A Le, Hoang Quynh %A Nguyen, Minh Trang %A Vu, Tien Sinh %D 2018 %F SisLab:2970 %I Oxford University Press %J Bioinformatics %T D3NER: Biomedical named entity recognition using CRF-biLSTM improved with fine-tuned embeddings of various linguistic information %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2970/ %0 Conference Paper %A Dang, Thanh Hai %A Lee, Wonjun %A Ha, Quang Thuy %B IEEE International Conference on Networking, Sensing and Control (ICNSC) %D 2008 %F SisLab:2079 %P 1496-1499 %T A pageranking based method for identifying characteristic genes of a disease %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2079/ %0 Journal Article %@ 0866-8612 %A Dang, Thanh Hai %A Nguyen, Dai Thanh %A Pham, Thi Minh Trang %A Dang, Cao Cuong %A Hoang, Kim Phuc %A Pham, Bao Son %A Le, Sy Vinh %A Le, Si Quang %A Phan, Thi Thu Hang %A Do, Duc Dong %D 2014 %F SisLab:2086 %I VNU Journal of Science %J VNU Journal of Computer Science and Communication Engineering %N 3 %P 31-35 %T Preliminary results on the whole genome analysis of a Vietnamese individual %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2086/ %V 30 %X We present some preliminary results on the whole genome analysis of an anonymous Kinh Vietnamese (KHV) individual that was deeply sequenced to 34-fold using the Illumina sequencing technology. The sequenced genome covered 99.85% of the human reference genome (GRCh37). We discovered (1) 3408825 single polymorphism nucleotides (SNPs) of which 41396 (1.2%) were novel, (2) 654024 short indels of which 35263 (5.4%) were novel, i.e. not present in the dbSNP and 1000 genomes project databases. We also detected 10611 large SVs (the length ≥ 100bp) of four types, i.e. large indels (90.6%), the inter-chromosomal translocations (3.1%), the inversions (3.4%) and the intra-chromosomal translocations (2.9%). This study is our initial step toward large-scale projects on Vietnamese population. %0 Journal Article %@ 0250-5991 %A Dang, Thanh Hai %A Nguyen, Dai Thanh %A Pham, Thi Minh Trang %A Le, Si Quang %A Phan, Thi Thu Hang %A Dang, Cao Cuong %A Hoang, Kim Phuc %A Nguyen, Huu Duc %A Do, Duc Dong %A Bui, Quang Minh %A Pham, Bao Son %A Le, Sy Vinh %D 2015 %F SisLab:1476 %I Springer %J Journal of Bioscience %N 1 %P 114-123 %T Whole Genome Analysis of a Vietnamese Trio %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1476/ %V 40 %X We here present the first whole genome analysis of an anonymous Kinh Vietnamese (KHV) trio whose genomes were deeply sequenced to 30-fold average coverage. The resulting short reads covered 99.91% of the human reference genome (GRCh37d5). We identified 4,719,412 SNPs and 827,385 short indels that satisfied the Mendelian inheritance law. Among them, 109,914 (2.3%) SNPs and 59,119 (7.1%) short indels were novel. We also detected 30,171 structural variants of which 27,604 (91.5%) were large indels. There were 6,681 large indels in the range 0.1–100 kbp occurring in the child genome that were also confirmed in either the father or mother genome. We compared these large indels against the DGV database and found that 1,499 (22.44%) were KHV specific. De novo assembly of high-quality unmapped reads yielded 789 contigs with the length ≥300 bp. There were 235 contigs from the child genome of which 199 (84.7%) were significantly matched with at least one contig from the father or mother genome. Blasting these 199 contigs against other alternative human genomes revealed 4 novel contigs. The novel variants identified from our study demonstrated the necessity of conducting more genome-wide studies not only for Kinh but also for other ethnic groups in Vietnam. %0 Journal Article %@ 1460-2059 %A Dang, Thanh Hai %A Van Leemput, Koenraad %A Verschoren, Alain %A Laukens, Kris %D 2008 %F SisLab:2073 %I Oxford University Press %J Bioinformatics %N 24 %P 2857-2864 %T Prediction of kinase-specific phosphorylation sites using conditional random fields %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2073/ %V 24 %0 Conference Paper %A Dang, Thi Huong Giang %A Vuong, Quang Huy %A Pham, Minh Trien %B 2018 REV-ECIT: National Conference on Electronics, Communications and Information Technology %C Hanoi, Vietnam %D 2018 %F SisLab:3245 %P 32-37 %T Cải Tiến Thuật Toán Tối Ưu Hoá Bầy Đàn Cho Bài Toán Lập Quỹ Đạo Bay Của UAV Trong Không Gian Ba Chiều %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3245/ %X Máy bay không người lái được sự quan tâm lớn trong các ứng dụng nông nghiệp thông minh, giám sát chất lượng công trình, hỗ trợ tìm kiếm cứu hộ cứu nạn... và đặc biệt là ứng dụng trong quân sự. Trong bài báo cáo này, chúng tôi tập trung nghiên cứu vấn đề lập kế hoạch bay cho UAV trong không gian 3-D biết trước. Chúng tôi sử dụng giải thuật là tối ưu bầy đàn (PSO) cải tiến để tối ưu quỹ đạo chuyển động của UAV, đồng thời, so sánh với giải thuật PSO truyền thống và giải thuật di truyền GA để thấy được tính ưu việt của PSO cải tiến. Quỹ đạo tối ưu của UAV có cánh cố định được định nghĩa như một hàm đa mục tiêu bao gồm các đoạn thẳng, đường cong và độ cao. Các giải thuật được phát triển và so sánh thông qua bản đồ thực tế của một số tỉnh ở Việt Nam. Kết quả ban đầu cho thấy tiềm năng áp dụng các giải thuật này trong việc tối ưu quỹ đạo bay cho UAV. Về tương lai, các giải thuật này sẽ tiếp tục được cải tiến nhằm giảm thời gian tối ưu hướng đến bài toán lập quỹ đạo bay thời gian thực cho UAV %0 Conference Paper %A Dang, Tran Thai %A Nguyen, Thi Xuan Huong %A Le, Anh Cuong %A Huynh, Van Nam %B 2014: The 6th International Conference on Knowledge and Systems Engineering (KSE 2014) %C Hanoi, Vietnam %D 2014 %F SisLab:448 %P 629-640 %T Automatically Learning Patterns in Subjectivity Classification for Vietnamese %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/448/ %0 Journal Article %@ 1380-7501 %A Dang, Trung Kien %A Worring, Marcel %A Bui, The Duy %D 2015 %F SisLab:1490 %I Springer %J Multimedia Tools and Applications %N 13 %P 4617-4639 %T Building 3D Event Logs for Video Investigation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1490/ %V 74 %X In scene investigation, creating a video log captured using a handheld camera is more convenient and more complete than taking photos and notes. By introducing video analysis and computer vision techniques, it is possible to build a spatio-temporal representation of the investigation. Such a representation gives a better overview than a set of photos and makes an investigation more accessible. We develop such methods and present an interface for navigating the result. The processing includes (i) segmenting a log into events using novel structure and motion features making the log easier to access in the time dimension, and (ii) mapping video frames to a 3D model of the scene so the log can be navigated in space. Our results show that, using our proposed features, we can recognize more than 70 percent of all frames correctly, and more importantly find all the events. From there we provide a method to semi-interactively map those events to a 3D model of the scene. With this we can map more than 80 percent of the events. The result is a 3D event log that captures the investigation and supports applications such as revisiting the scene, examining the investigation itself, or hypothesis testing. %0 Conference Paper %A Dang, Van Do %A Hoang, Xuan Tung %A Kim-Khoa, Nguyen %B IEEE Globecom 2020 %D 2020 %F SisLab:4251 %T Optimization of IoT Service Deployment In Multi-Layered Cloud-Fog Environment %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4251/ %X Recently, fog computing, which can be done in proximity to data sources, has emerged as a solution to provide low-latency Quality-of-Service (QoS) for IoT services in complement to centralized cloud with unlimited computing resources. Optimized service deployment on both cloud and fog environments is challenging due to their heterogeneity. Prior works mainly focus on mapping service functions and dependencies directly to physical network. In this paper, we propose a multi-layer mapping mechanism that efficiently deploys multiple IoT services to the appropriate virtual networks in physical infrastructure. We design greedy-based algorithms for solving this NP-hard problem with two phases executed sequentially. Experimental results show our proposed solution can reduce upto 80% of the total service cost compared to the state-of-the-art solutions. %0 Conference Paper %A Dang, Van Do %A Hoang, Xuan Tung %A Tran, Mai %A Nguyen, Kim Khoa %B IEEE Global Communications Conference %D 2020 %F SisLab:4211 %T Optimization of IoT Service Deployment In Multi-Layered Cloud-Fog Environment %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4211/ %0 Journal Article %A Dao, Dzung Viet %A Nakamura, Koichi %A Tung, Bui Thanh %A Sugiyama, Susumu %D 2010 %F SisLab:209 %J Advances in Natural Sciences: Nanoscience and Nanotechnology %N 1 %P 013001 %T Micro/nano-Mechanical Sensors and Actuators based on SOI-MEMS Technology %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/209/ %V 1 %X MEMS (micro-electro-mechanical systems) technology has undergone almost 40 years of development, with significant technology advancement and successful commercialization of single-functional MEMS devices, such as pressure sensors, accelerometers, gyroscopes, microphones, micro-mirrors, etc. In this context of MEMS technology, this paper introduces our studies and developments of novel micro/nano-mechanical sensors and actuators based on silicon- on-insulator (SOI)-MEMS technology, as well as fundamental research on piezoresistive effects in single-crystal silicon nanowires (SiNWs). In the first area, novel mechanical sensors, such as 6-DOF micro-force moment sensors, multi-axis inertial sensors and micro-electrostatic actuators developed with SOI-MEMS technology will be presented. In the second area, we have combined atomic-level simulation and experimental evaluation methods to explain the giant piezoresistive effect in single crystalline SiNWs along different crystallographic orientations. This discovery is significant for developing more highly sensitive and miniaturized mechanical sensors in the near future. %0 Conference Paper %A Dao, Ngoc Thanh %A Bui, Quang Hung %A Nguyen, Hai Chau %A Nguyen, Van Vuong %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1518 %T Building the Local Spatial Data Infrastructure for the Northeast and Northwest of Vietnam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1518/ %X National Spatial Data Infrastructure (NSDI) is the backbone of the development of e-government for a country. Since 2011, Vietnam began to pay attention to NSDI. Till now, the is no significant progress in implementation of NSDI except a common research between World Bank and Ministry of Natural Resources and Environment on strategic planning of NSDI and another research on making policy for building NSDI. In this paper, we present our research on the construction of the Local SDI (LSDI) for the Northeast and the Northwest of Vietnam. Our hypothesis is that the success of this LSDI will be a good lesson for building NSDI in the future. The reason why we selected the Northeast and Northwest because it is a large area with a total area of approximately 115,700km2 that is nearly 1/3 of Vietnam. Moreover, the Northeast and the Northwest comprehensively contain of topographical and morphological characteristics of nature, economics and society of Vietnam. NSDI is considered consisting of seven components: (i) spatial data; (ii) standards; (iii) Access to data; (iv) technology; (v) policy – legislation; (vi) organization – institution; (vii) partnership. In this research we only focused on four technical components which are spatial data, standards, access to data and technology. We collected spatial data from three sources: (i) existing spatial data and framework spatial data from Government and Local Government; (ii) existing remote sensing data from the NASA, USGS; (iii) Update remote sensing data from MODIS/NPP ground station at our university. We used open source platforms for building LSDI. Regarding to the access to data, we have organized spatial data in a data warehouse architecture to develop BI (Business Intelligence) tools. A pilot tool is for automatically generating annual reports on 43 indicators of sustainable development issued by the Prime Minister. For the future work, we will evaluate this LSDI from the viewpoints of end-users from government, local government and citizens. %0 Conference Paper %A Dao, Thi Huong %A Le, Hong Anh %A Truong, Ninh Thuan %B 5th EAI International Conference on Context-Aware Systems and Applications %C Vietnam %D 2016 %F SisLab:1902 %T An approach to analyzing execution preservation in Java program refactoring %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1902/ %0 Conference Paper %A Dao, Thi Huong %A Le, Hong Anh %A Truong, Ninh Thuan %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1519 %T An approach to verifying consistency of software refactoring process %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1519/ %X In software development, refactoring is a process that improves the system internal structure without altering its external behavior. Applying design patterns, which is the common reusable solution of several kinds of problems, is widely adopted in refactoring. And, the combination of refactoring and design patterns has established a revolution approach in software re-engineering. It, however, raises a challenged issue that the software model may not preserve some cer- tain behavioral properties. In this paper, we propose a new approach to checking the consistency between original model and its evolution (the model after applying design pattern). To this aim, we employ Unified Modeling Language (UML) to modeling system, and Object Constraint Language (OCL) to represent the system requirement. In additional, we construct a set of consistent rule which uses to checking whether the system can preserve its behavior or not. We also illus- trate the checking approach in a case study of Adaptive Road Traffic Control System. %0 Conference Paper %A Dao, Thi Huong %A To, Van Khanh %A Truong, Ninh Thuan %A Chu, Quang Ton %A Nguyen, Viet Ha %B NICS: 2nd National Foundation for Science and Technology Development Conference on Information and Computer Science %C Ho Chi Minh city, Vietnam %D 2015 %F SisLab:1465 %K Unified Modeling Language;authorisation;data encapsulation;object-oriented methods;UML model;access control mechanisms;access permission constraints;encapsulation;object oriented models;structured data objects;Access control;Encapsulation;Medical services;Object oriented modeling;Software;Unified modeling language;Access control;UML class diagram;specification;verification %P 49-53 %T Integrating and Checking Access Permissions in Object Oriented Models %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1465/ %X Encapsulation in object oriented models refers to the bundling of data with the methods that operate on that data, it is used to hide the values or state of a structured data objects and preventing unauthorized parties' direct access to them. Access control mechanism in information systems is the process of mediating every request to resources and determining whether the request should be granted or denied. This paper proposes an approach allowing us to integrate access control mechanisms in object oriented paradigm. More detail, we define the access permissions between methods of classes in a UML model and providing a prototype to verify if the program implementation satisfied their access permission constraints. %0 Conference Paper %A Dao, Thi Huong %A Trinh, Thanh Binh %A Truong, Ninh Thuan %B KSE: the 2017 International Conference on Knowledge and Systems Engineering %C Hue, Vietnam %D 2017 %F SisLab:2655 %T A Tool Support for Checking Consistency in Model Refactoring %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2655/ %X Refactoring is a technique that transforms a software to new one in order to improve the internal structure without affect its external behaviors. Although its benefits, refactoring is not widely adopted in practice as it is attainable. One of the biggest challenges to refactoring has been the hapless lack of tool support that may be raised some issues such as tedious, time-consuming and error-prone. We present in this paper a tool called CVT (Consistency Validator Tool) that used to support to checking consistency in model refactoring. The CVT is composed of three essential functions which are well-formedness checking, translate OCL to FOL and behavioral consistency checking. We are most interested in OCL to FOL translation process that make use of Abstract Syntax Tree (AST) as an intermediate syntax. We have also carried out some experimental studies in order to evaluate the effectiveness of this tool. %0 Conference Paper %A Dinh, Bao Minh %A Duong, Van Minh %A Nguyen, Quoc Bao %A Ngo, Sy Trung %A Nguyen, Truong Son %B Hội nghị Quốc gia lần thứ XXIV về Điện tử, Truyền thông và Công nghệ Thông tin %F SisLab:4662 %T Trợ lý ảo cho Robot tiếp tân Khách sạn Dựa trên Lý thuyết sRAM và nền tảng Dialogflow %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4662/ %X Robot ngày càng trở nên phổ biến trong ngành dịch vụ và du lịch. Đặc biệt, trong bối cảnh của dịch COVID-19, robot tiếp tân đã trở thành giải pháp giúp hạn chế tiếp xúc và giảm nguy cơ lây nhiễm. Là dịch vụ tư vấn và hướng dẫn khách hàng, công việc tiếp tân đòi hỏi robot phải có khả năng tương tác linh hoạt với khách hàng. Bài báo trình bày một trợ lý ảo hỗ trợ robot tiếp tân khách sạn tương tác với khách hàng dựa trên cơ sở lý thuyết “Service Robot Acceptance Model" (sRAM). Trợ lý ảo được xây dựng trên nền tảng Dialogflow của Google. Kết quả kiểm thử với nhiều kịch bản hội thoại cho thấy hệ thống có độ chính xác cao, có độ trễ thấp và có thể áp dụng trong thực tế. Một phần kết quả thực nghiệm được ghi hình và trình bày tại: Kết quả thực nghiệm %0 Conference Paper %A Dinh, Bao Minh %A Tran, Duc Son %A Nguyen, The Huong %A Do, Nam %A Le, Minh Hoang %A Hoang, Van Xiem %B Hội nghị Quốc gia lần thứ XXIV về Điện tử, Truyền thông và Công nghệ Thông tin %F SisLab:4661 %T MAI_ARM: Robot Tay Máy Thông Minh sử dụng Trí tuệ nhân tạo Đa thể thức %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4661/ %X Các hệ thống robot tay máy đóng một vai trò rất quan trọng trong sản xuất công nghiệp hiện đại. Việc ứng dụng robot tay máy trong đời sống hàng ngày sẽ giúp ích con người rất nhiều, ví dụ như tự động hóa hoàn toàn quy trình phân phát thuốc và thực phẩm cho bệnh nhân nhiễm COVID-19. Ứng dụng robot tay máy trong đời sống đòi hỏi một cách thức giúp người dùng dễ dàng tương tác với robot. Trong bài báo này, một phương pháp tương tác với robot tay máy sử dụng trí tuệ nhân tạo đa thể thức kết hợp giọng nói-hình ảnh được xây dựng. Bài báo cũng đề xuất phương pháp Chessboard Calibration giúp nâng cao độ chính xác trong việc xác định vị trí thực thi của robot tay máy. Một robot tay máy 4 bậc tự do (4-DOF) chế tạo bằng phương pháp in 3D sẽ được sử dụng để thực thi và đánh giá mô hình trí tuệ nhân tạo đa thể thức xây dựng được. Kết quả thực thi hệ thống được ghi hình và trình bày tại: https://youtu.be/RHgAyHXMH6I %0 Conference Paper %A Dinh, Ngoc Thi %A Vo, Dinh Hieu %A Vu, Thi Dao %A Nguyen, Viet Ha %B The 9th Asian Conference on Intelligent Information and Database Systems %C Kanazawa, Japan %D 2017 %F SisLab:2501 %I Springer International Publishing %P 499-513 %T Generation of Test Data Using Genetic Algorithm and Constraint Solver %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2501/ %0 Conference Paper %A Dinh, Phu Hung %A Nguyen, Ngoc Khuong %A Le, Anh Cuong %A VNU-UET, %B 2012 IEEE RIVF International Conference on Computing and Communication Technologies, Research, Innovation, and Vision for the Future (RIVF) %C Ho Chi Minh city %D 2012 %F SisLab:112 %K Accuracy;Context;Data models;Learning systems;Machine learning;Niobium;Training;learning (artificial intelligence);natural language processing;statistical analysis;Vietnamese word sense disambiguation;general statistical model;statistical machine learning model;transformation based learning;transformation rule learning; %P 1 -6 %T Combining Statistical Machine Learning with Transformation Rule Learning for Vietnamese Word Sense Disambiguation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/112/ %X Word Sense Disambiguation (WSD) is the task of determining the right sense of a word depending on the context it appears. Among various approaches developed for this task, statistical machine learning methods have been showing their advantages in comparison with others. However, there are some cases which cannot be solved by a general statistical model. This paper proposes a novel framework, in which we use the rules generated by transformation based learning (TBL) to improve the performance of a statistical machine learning model. This framework can be considered as a combination of a rule-based method and statistical based method. We have developed this method for the problem of Vietnamese WSD and achieved some promising results. %0 Book Section %A Dinh, Quang Huy %A Ma, Thi Chau %A Bui, The Duy %A Nguyen, Trong Toan %A Nguyen, Dinh Tu %B New Challenges for Intelligent Information and Database Systems %D 2011 %F SisLab:2653 %I SpringerLink %P 117-126 %T Facial soft tissue thicknesses prediction using anthropometric distances %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2653/ %X Predicting the face of an unidentified individual from its skeletal remains is a difficult matter. Obviously, if the soft tissue thicknesses at every location at the skull are known, we can easily rebuild the face from the skull model. Thus, the problem turns out to be predicting the soft tissue thicknesses for any given skull. With the rapid development of the computer, different techniques are being used in the community for prediction tasks and in recent years the concept of neural networks has emerged as one of them. The principal strength of the neural network is its ability to find patterns and irregularities as well as detecting multi-dimensional non-linear connections in data. In this paper, we propose a method of applying neural networks to predict the soft tissue thicknesses for facial reconstruction. We use the distances between anthropometric locations at the skull as input, and the soft tissue thicknesses as output, as this format is suitable for many machine learning mechanisms. These data is collected and measured from candidates using the Computed Tomography (CT) technique. %0 Conference Paper %A Dinh, Thi Ngoc %A Vo, Dinh Hieu %A Nguyen, Viet Ha %B The 2016 International Conference on Advanced Computing and Applications %C Vietnam %D 2016 %F SisLab:2061 %T A Technique for Generating Test Data using Genetic Algorithm %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2061/ %0 Journal Article %@ 978-1-6654-7251-7 %A Dinh, Tran Hiep %A Nguyen, Linh Trung %D 2022 %F SisLab:4775 %I IEEE %J 2022 IEEE 16th International Symposium on Medical Information and Communication Technology (ISMICT) %P 1-6 %T Masked Face Detection with Illumination Awareness %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4775/ %X Mask mandate has been applied in many countries in the last two years as a simple but effective way to limit the Covid-19 transmission. Besides the guidance from authorities regarding mask use in public, numerous vision-based approaches have been developed to aid with the monitoring of face mask wearing. Despite promising results have been obtained, several challenges in vision-based masked face detection still remain, primarily due to the insufficient of a quality dataset covering adequate variations in lighting conditions, object scales, mask types, or occlusion levels. In this paper, we investigate the effectiveness of a lightweight masked face detection system under different lighting conditions and the possibility of enhancing its performance with the employment of an image enhancement algorithm and an illumination awareness classifier. A dataset of human subjects with and without face masks in different lighting conditions is first introduced. An illumination awareness classifier is then trained on the collected dataset, the labeling of which is processed automatically based on the difference in detection accuracy when an image enhancement algorithm is taken into account. Experimental results have shown that the combination of the masked face detection system with the illumination awareness and an image enhancement algorithm can boost the system performance to up to 8.6%, 7.4%, and 8.5% in terms of Accuracy, F1-score, and AP-M, respectively. %0 Journal Article %@ 2287-5255 %A Dinh, Trieu Duong %A Le Dinh, Minh %A Byeungwoo, Jeon %A Hoang, Van Xiem %D 2019 %F SisLab:3454 %I The Institute of Electronics and Information Engineers (IEIE) %J IEIE Trans. on Smart Processing and Computing %T A Novel Fusion Method for 3D-TV View Synthesis Using Temporal and Disparity Correlations %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3454/ %X View synthesis like Depth-image-based-rendering (DIBR) plays a significant role in 3D content creation for 3D-TV. However, perceptual errors introduced by current view synthesis often result in severe distortions in synthesized images. In this paper, we propose a novel view synthesis fusion (VSF) method which adaptively exploits temporal and disparity correlations to improve the quality of the synthesized picture. The proposed VSF method defines a robust correlation assessment metric for fusing several pre-created virtual view candidates. Unlike conventional methods, the proposed fusion algorithm is applied for both hole and non-hole areas. Experimental results show significantly outperforming peak signal-to noise ratio (PSNR) and subjective visual quality by the proposed method compared to other conventional methods. %0 Journal Article %@ 1999-4893 %A Dinh, Trieu Duong %A Phi, Cong Huy %A Hoang, Van Xiem %D 2019 %F SisLab:3455 %I MDPI %J MDPI algorithms %P 1-10 %T A Novel Consistent Quality Driven for JEM based Distributed Video Coding %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3455/ %X Distributed video coding (DVC) is an attractive and promising solution for low complexity constrained video applications, such as wireless sensor networks or wireless surveillance systems. In DVC, the visual quality consistency is one of the most important issues to evaluate the performance of a DVC codec. However, it is the fact that the quality of the decoded frames achieved in most recent DVC codecs is not consistent and it is varied with high quality fluctuation. To solve the problem, in this paper, we propose a novel DVC solution named JEM based DVC (JEM-DVC), which can provide not only higher performance compared to the traditional DVC solutions but also effective scheme for the quality consistency control. In the proposed JEM-DVC solution, we first employ several advanced techniques provided in the Joint exploration model (JEM) of the future video coding standard (FVC) to effectively improve the performance of JEM-DVC codec. Then, for consistent quality control, we propose two novel methods named key frame quantization (KF-Q) and Wyner-Zip frame quantization (WZF-Q) which determine the optimal values of quantization parametter (QP) and quantization matrix (QM) applied for the key and WZ frame coding, respectively. Unlike the conventional approaches, the optimal values of QP and QM are adaptively controlled and updated for every key and WZ frames to guarantee the consistent video quality for the proposed codec. Our proposed JEM-DVC is the first DVC codec in literature employing JEM coding technique, then all results presented in this paper are new. Experimental results show that the proposed JEM-DVC significantly outperforms the relevant DVC benchmarks, notably the DISCOVER DVC and the recent H.265/HEVC based DVC, in terms of both Peak signal-to-noise ratio (PSNR) performance and consistent visual quality. %0 Journal Article %@ 2196-8888; Online 2196-8896 %A Divrood, Ali Rezaei %A Ha, Quang Thuy %A Nguyen, Linh Anh %A Nguyen, Hung Son %D 2017 %F SisLab:2638 %I Springer Berlin Heidelberg %J Vietnam Journal of Computer Science %P 1-12 %T On the possibility of correct concept learning in description logics %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2638/ %V 2017 %X It is well known that any Boolean function in classical propositional calculus can be learned correctly if the training information system is good enough. In this paper, we extend that result for description logics. We prove that any concept in any description logic that extends ALCALC with some features amongst I (inverse roles), QkQk (qualified number restrictions with numbers bounded by a constant k), and SelfSelf (local reflexivity of a role) can be learned correctly if the training information system (specified as a finite interpretation) is good enough. That is, there exists a learning algorithm such that, for every concept C of those logics, there exists a training information system such that applying the learning algorithm to it results in a concept equivalent to C. For this result, we introduce universal interpretations and bounded bisimulation in description logics and develop an appropriate learning algorithm. We also generalize common types of queries for description logics, introduce interpretation queries, and present some consequences. %0 Conference Paper %A Divroodi, Ali Rezaei %A Ha, Quang Thuy %A Nguyen, Linh Anh %A Nguyen, Hung Son %B The 4th International Conference on Computational Collective Intelligence (ICCCI 2012) %C Ho Chi Minh, Vietnam %D 2012 %F SisLab:80 %P 230-238 %T On C-Learnability in Description Logics %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/80/ %0 Journal Article %@ 2196-8888 %A Divroodi, Ali Rezaei %A Ha, Quang Thuy %A Nguyen, Linh Anh %A Nguyen, Hung Son %A Education DepartmentNoshahrIran, %A Faculty of Information TechnologyVNU University of Engineering and TechnologyHanoiVietnam, %A Division of Knowledge and System Engineering for ICTTon Duc Thang UniversityHo Chi Minh CityVietnam, %A Institute of InformaticsUniversity of WarsawWarsawPoland, %D 2018 %F SisLab:3154 %I Vietnam Journal of Computer Science %J Vietnam Journal of Computer Science %N 1 %P 3-14 %T On the possibility of correct concept learning in description logics %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3154/ %V 5 %X It is well known that any Boolean function in classical propositional calculus can be learned correctly if the training information system is good enough. In this paper, we extend that result for description logics. We prove that any concept in any description logic that extends ALC with some features amongst I (inverse roles), Qk (qualified number restrictions with numbers bounded by a constant k), and Self (local reflexivity of a role) can be learned correctly if the training information system (specified as a finite interpretation) is good enough. That is, there exists a learning algorithm such that, for every concept C of those logics, there exists a training information system such that applying the learning algorithm to it results in a concept equivalent to C. For this result, we introduce universal interpretations and bounded bisimulation in description logics and develop an appropriate learning algorithm. We also generalize common types of queries for description logics, introduce interpretation queries, and present some consequences. %0 Journal Article %@ 2301-3702 %A Do, Anh Tuan %A Ngo, Thi Duyen %A Bui, The Duy %D 2016 %F SisLab:1981 %J Journal of Automation and Control Engineering %N 2 %P 171-176 %T Feature Extraction for Non-frontal Faces %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1981/ %V 4 %X One of the most challenge tasks in building a face recognition system is how to represent and extract good quality features from face images. The difficulties come from variations in head poses, illumination conditions, and facial expression. Although many researches have been done, most were carried on under constrained environments. Most researches concentrated on dealing with frontal faces. Processing non-frontal faces encounters more challenge because some features on faces become occluded dramatically. In this paper, we propose two models to extract features from non-frontal faces in the range of 30o to 90o. First, we use the Viola-Jones detection method to identify the pose of face images. Then, we use Active Appearance Model (AAM) to interpret face images. Lastly, the models are trained to know how to fit new images. To improve the efficiency of fitting, we apply a nonlinear parameter update method. Experimental results show that using nonlinear fitting for non-frontal can increase the accuracy of the AAM fitting, compared with some previous methods. %0 Report %9 Technical Report %@ TR2017-FIT-002 %A Do, Duc Dong %D 2017 %F SisLab:2894 %T A Novel and Efficient Ant Colony Optimization Algorithm for Protein 3D Structure Prediction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2894/ %X Protein structure prediction (PSP) is considered as one of the most long-standing and challenging problem in bioinformatic. In this paper, we present an efficient ant colony optimization algorithm to predict the protein structure on three-dimensional face-centered cubic lattice coordinates, using hydrophobic-polar (HP) model and MiyazawaJernigan (MJ) model to calculate the free energy. The reinforcement learning information is expressed in the k-order Markov model, the heuristic information is determined based on the increase of the total energy. On a set of benchmark proteins, the results show a remarkable efficiency of our algorithm by comparing with the state-of-the-art algorithms. %0 Conference Paper %A Do, Duc Dong %A Dinh, Quang Huy %A Dang, Thanh Hai %A Laukens, Kris %A Hoang, Xuan Huan %B International Conference on Swarm Intelligence %D 2012 %F SisLab:2080 %P 204-211 %T AcoSeeD: An ant colony optimization for finding optimal spaced seeds in biological sequence search %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2080/ %0 Conference Paper %A Do, Duc Dong %A Thai Dinh, Phuc %A Vu, Thi Ngoc Anh %A Nguyen, Linh Trung %B 12th International Symposium on Medical Information and Communication Technology (ISMICT) %C Sydney, Australia %D 2018 %F SisLab:2948 %P 28-33 %T An Efficient Ant Colony Optimization Algorithm for Protein Structure Prediction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2948/ %X Protein structure prediction is considered as one of the most long-standing and challenging problem in bioinformatics. In this paper, we present an efficient ant colony optimization algorithm to predict the protein structure on three-dimensional face-centered cubic lattice coordinates, using the hydrophobic–polar model and the Miyazawa-Jernigan model to calculate the free energy. The reinforcement learning information is expressed in the k-order Markov model, and the heuristic information is determined based on the increase of the total energy. On a set of benchmark proteins, the results show a remarkable efficiency of our algorithm in comparison with several state-of-the-art algorithms. %0 Conference Paper %A Do, Duc Dong %A Tran, Ngoc Ha %A Dang, Thanh Hai %A Dang, Cao Cuong %A Hoang, Xuan Huan %B The 2015 International Conference Advanced Technologies for Communications (ATC) %C Ho Chi Minh city, Vietnam %D 2015 %F SisLab:1597 %P 332 %T An efficient algorithm for global alignment of protein-protein interaction networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1597/ %X Global aligning two protein-protein interaction networks is an essentially important task in bioinformatics computational biology field of study. It is a challenging and widely studied research topic in recent years. Accurately aligned networks allow us to identify functional modules of proteins and/ororthologous proteins from which unknown functions of a protein can be inferred. We here introduce a novel efficient heuristic global network alignment algorithm called FASTAn, including two phases: the first to construct an initial alignment and the second to improve such alignment by exerting a local optimization repeated procedure. The experimental results demonstrated that FASTAn outperformed the stateof-the-art global network alignment algorithmnamely SPINAL in terms of both commonly used objective scoresand the run-time. Keywords: FASTAn, Heuristic algorithm, Biological network alignment, Protein-protein interaction networks %0 Conference Paper %A Do, Khac Phong %A Ma, Thi Chau %A Cao, Hoang Giang %A Nguyen, Thi Thu An %B 2018 The 5th NAFOSTED Conference on Information and Computer Science (NICS) %C Ho Chi Minh city, Vietnam %D 2018 %F SisLab:3122 %T Region-based deformation transfer %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3122/ %X Mesh deformation is a fundamental technique for geometric modeling which is applied successfully in a wide range of applications from shape design to computer animation. Normally, the deformation transferred from one actor to another actor is based on all vertices and triangles of a mesh, leading to timeconsuming in terms of a 3D sequential model. To address this problem, we propose a region-based deformation transfer that automatically identifies several regions with the largest displacement in time series, and then exploits those deformations of such regions. Our experimental results demonstrate that we can obtain the similar deformed mesh in spite of using approximately 50% – 60% of the facial area, therefore the time decrease significantly for deformation transfer step. %0 Journal Article %@ 1976-913X %A Do, Khac Phong %A Nguyen, Ba Tung %A Nguyen, Xuan Thanh %A Bui, Quang Hung %A Tran, Nguyen Le %A Nguyen, Thi Nhat Thanh %A Vuong, Van Quynh %A Le, Thanh Ha %D 2015 %F SisLab:1480 %I KIPS %J Journal of Information Processing Systems %T Spatial Interpolation and Assimilation Methods for Satellite and Ground Meteorological Data in Vietnam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1480/ %0 Conference Paper %A Do, Thuy %A Pham, Hoang Giang %A Tran, Xuan Khoi %A Ha, Minh Hoang %A Do, Bach %B The 4th NAFOSTED Conference on Information and Computer Science (NICS) %C Hanoi, Vietnam %D 2017 %F SisLab:2761 %T Solving the staff rescheduling problem in Lai Chau hydropower station %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2761/ %X This article studies the complex staff rescheduling problem arising in the context of Lai Chau hydropower station. Currently, rescheduling task is implemented manually by agreement of switching shifts among workers or by a scheduler. It is a complex and time-consuming operation but obtained schedules have been far below the expectation of the staff as they in general violate working rules and contract agreements. In this research, we propose an approach based on Mixed Integer Programming to support the scheduler solving the problem. Our model includes multiple objectives which minimize the number of changed shifts and the number of affected employees. Experimental results show the effectiveness of our method. %0 Conference Paper %A Do, Van Duy %A To, Van Khanh %B SSC: the 2016 Student Scientific Contest %C Hanoi %D 2016 %F SisLab:1501 %T SAT Encoding dựa trên mẫu tìm kiếm giải trò chơi Loogic Hitori %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1501/ %0 Journal Article %@ 1935-3812 %A Do Duc, Dong %A Le, Sy Vinh %A Hoang, Xuan-Huan %D 2013 %F SisLab:186 %J Swarm Intelligence %N 1 %P 63-77 %T ACOHAP: An Efficient Ant Colony Optimization for the Haplotype Inference by Pure Parsimony Problem %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/186/ %V 7 %0 Journal Article %A Do Duc, Dong %A Le, Sy Vinh %A Hoang, Xuan-Huan %D 2013 %F SisLab:184 %J Swarm Intelligence %N 1 %P 63-77 %T ACOHAP: An Efficient Ant Colony Optimization for the Haplotype Inference by Pure Parsimony Problem %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/184/ %V 7 %0 Journal Article %A Doan, Quang Manh %A Dinh, Tran Hiep %A Nguyen, Linh Trung %A Nguyen N, Diep %A Singh, Avinash Kumar %A Lin, Chin-Teng %F SisLab:4817 %J To appear in the 2023 IEEE Statistical Signal Processing Workshop (SSP) proceedings %T Extended Upscale and Downscale Representation with Cascade Arrangement %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4817/ %X Smoothing filters are widely used in EEG signal processing for noise removal while preserving important features. Unlike common approaches in the time domain, a recent effective algorithm using the Upscale and Downscale Representation (UDR) technique has been introduced to process the signal in the image domain. The idea of UDR is to visualize the input with an appropriate line width, convert it to a binary image, and then smooth it by skeletonizing the signal object to a unit width and projecting it back to the time domain. We propose in this paper a cascaded UDR (CUDR) where the interested signal is filtered twice. CUDR’s performance is verified on simulated data with added white Gaussian noise and compared with the cascaded arrangement of some conventional techniques. Experimental results have demonstrated the outperformance of CUDR in terms of the fitting error when dealing with noisy signals, especially at a low signal-to-noise ratio. %0 Conference Paper %A Doan, Van Viet %A Nguyen, Duy Hung %A Tran, Quoc Long %A Nguyen, Do Van %A Le, Thanh Ha %B 2018 Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems in conjunction with Intelligent Systems Workshop %C Toyama, Japan %D 2018 %F SisLab:3263 %T Real-time Image Semantic Segmentation Networks with Residual Depth-wise Separable Blocks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3263/ %X —Semantic image segmentation plays a key role in obtaining pixel-level understanding of images. In recent years, researchers have tackled this problem by using deep learning methods instead of traditional computer vision methods (eg [25]). Because of the development of technologies like autonomous vehicles and indoor robots, segmentation techniques, that have not only high accuracy but also the capability of running in real-time on embedded platform and mobile devices, are in high demand. In this work, we have proposed a new convolutional module, named Residual depth-wise separable, and a fast and efficient convolutional neural network for segmentation. The proposed method is compared against other state of the art real-time models. The experiment results illustrate that our method is efficient in computation while achieves state of the art performance in term of accuracy %0 Conference Paper %A Doan, Van Viet %A Nguyen, Duy Hung %A Tran, Quoc Long %A Nguyen, Do Van %A Le, Thanh Ha %B Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems in conjunction with Intelligent Systems Workshop 2018 (SCIS&ISIS2018) %D 2018 %F SisLab:3293 %T Real-time Image Semantic Segmentation Networks with Residual Depth-wise Separable Blocks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3293/ %0 Journal Article %@ 2249 – 8958 %A Du, Phuong Hanh %A Duong, Ngoc Son %A Nguyen, Ngoc Cuong %A Nguyen, Ngoc Hoa %D 2020 %F SisLab:3990 %I Research Trend %J International Journal on Emerging Technologies %N 2 %P 370-377 %T A Fast Computation of the Betweenness Centrality in Social Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3990/ %V 11 %0 Conference Paper %A Du, Phuong Hanh %A Nguyen, Hai Chau %A Nguyen, Kim Khoa %A Nguyen, Ngoc Hoa %B 2018 the 9th International Symposium on Information and Communication Technology (SoICT) %C Da Nang City, Viet Nam %D 2018 %F SisLab:3479 %T An Efficient Parallel Algorithm for Computing the Closeness Centrality in Social Network %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3479/ %X Closeness centrality is an substantial metric used in large-scale network analysis, in particular social networks. Determining closeness centrality from a vertex to all other vertices in the graph is a high complexity problem. Prior work has a strong focuses on the algorithmic aspect of the problem, and little attention has been paid to the definition of the data structure supporting the implementation of the algorithm. Thus, we present in this paper an efficient algorithm to compute the closeness centrality of all nodes in a social network. Our algorithm is based on (i) an appropriate data structure for increasing the cache hit rate, and then reducing amount of time accessing the main memory for the graph data, and (ii) an efficient and parallel complete BFS search to reduce the execution time. We tested performance of our algorithm, namely BigGraph, with five different real-world social networks and compare the performance to that of current approaches including TeexGraph and NetworKit. Experiment results show that BigGraph is faster than TeexGraph and NetworKit 1.27-2.12 and 14.78-68.21 times, respectively. %0 Conference Paper %A Du, Phuong Hanh %A Nguyen, Hai Chau %A Nguyen, Kim Thoa %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1520 %T Nâng cao hiệu năng xử lý truy vấn với mô hình dữ liệu trong bộ nhớ %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1520/ %X Các hệ quản trị cơ sở dữ liệu (CSDL) truyền thống chủ yếu được xây dựng dựa trên việc khai thác các máy tính đơn nhân CPU và thiết bị lưu trữ kiểu từ tính. Các mô hình này chưa khai thác được hết những thế mạnh công nghệ của các chip đa nhân hay những hệ thống máy tính cụm và dung lượng càng ngày càng lớn của bộ nhớ chính. Để có thể nâng cao được hiệu năng xử lý truy vấn, nhiều mô hình xử lý mới đã được đưa ra như H-Store với tư tưởng dựa trên mô hình dữ liệu trong bộ nhớ chính; Map-Reduce kết hợp Hadoop FileSystem để xử lý phân tán, … Từ thực trạng đó, luận án này hướng đến mục tiêu nghiên cứu, cải thiện mô hình xử lý truy vấn với loại dữ liệu không gian chuyên biệt theo hai hướng kết hợp: khai thác tối đa không gian bộ nhớ chính khả dụng và phân rã các truy vấn để có thể tiến hành song song trên nhiều lõi/chip khác nhau. Mô hình thử nghiệm sẽ được tiến hành dựa trên những bài toán kinh điển như tìm lộ trình tối ưu giữa hai vị trí địa lý khác nhau, các truy vấn trên mạng xã hội, mạng IoT, … %0 Journal Article %@ 2190-9288 %A Du, Phuong Hanh %A Pham, Hai Dang %A Nguyen, Ngoc Hoa %D 2018 %F SisLab:2959 %I Springer %J Transactions on Computational Collective Intelligence %P 182-199 %T An Efficient Parallel Method for Optimizing Concurrent Operations on Social Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2959/ %V 29 %X This paper presents our approach to optimize the performance of both reading and writing concurrent operations on large-scale social network. Here, we focus on the directed, unweighted relationships among members in a social network. It can then be illustrated as a directed, unweighted graph. And determining the relationship between any two members is similar to finding the shortest path between two vertices. With such a large-scale dynamic social network, we face the problem of having concurrent operations from adding or removing edges dynamically while one may ask to determine the relationship between two members. To solve this issue, we propose an efficient parallel method based on (i) utilizing an appropriate data structure, (ii) parallelizing the updating actions and (iii) improving the performance of query processing by both reducing the searching space and computing in multi-threaded parallel. Our method was validated by the datasets from SigMod Contest 2016 and SNAP DataSet Collections with good experimental results compared to other solutions %0 Conference Paper %A Du, Phuong Hanh %A Pham, Hai Dang %A Nguyen, Ngoc Hoa %B ICCCI: Conference on Computational Collective Intelligence Technologies and Applications %D 2017 %E George, A. Papadopoulos %E Nguyen, Ngoc Thanh %F SisLab:2676 %I Springer %P 148-159 %T An Efficient Parallel Method for Performing Concurrent Operations on Social Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2676/ %V 10448 %X This paper presents our approach to optimize the concurrent operations on a large-scale social network. Here, we focus on the directed, unweighted relationships among members in a social network. It can then be illustrated as a directed, unweighted graph. With such a large-scale dynamic social network, we face the problem of having concurrent operations from adding or removing edges dynamically while one may ask to determine the relationship between two members. To solve this challenge, we propose an efficient parallel method based on (i) utilizing an appropriate data structure, (ii) optimizing the updating actions and (iii) improving the performance of query processing by both reducing the searching space and computing in multi-threaded parallel. Our method was validated by the datasets from SigMod Contest 2016 and SNAP DataSet Collections with the good experimental results compared to other solutions. %0 Conference Paper %A Du, Phuong Hanh %A Pham, Hai Dang %A Nguyen, Ngoc Hoa %B The 3rd IEEE/ACM International Conference on Big Data Computing, Applications and Technologies %D 2016 %F SisLab:2261 %P 210-216 %T Optimizing the Shortest Path Query on Large-Scale Dynamic Directed Graph %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2261/ %0 Journal Article %@ 0866-8612 %A Du, Phuong Hanh %A Pham, Hai Dang %A Vu, Ba Duy %D 2015 %F SisLab:1727 %I Vietnam National University Hanoi %J VNU Journal of Science %T Trực quan hoá dữ liệu lớn trong ứng dụng giám sát và điều khiển tập trung %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1727/ %0 Conference Paper %A Du, Phuong Hanh %A Vu, Ba Duy %A Nguyen, Ngoc Hoa %B Kỷ yếu Hội thảo Quốc gia lần thứ 9 về Nghiên cứu cơ bản và ứng dụng công nghệ thông tin (FAIR) %C Can Tho, Vietnam %D 2016 %F SisLab:1841 %T Trực quan hoá dữ liệu trạng thái giao thông trên nền Web %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1841/ %0 Conference Paper %A Duc, Dong Do %A Le, Tri-Thanh %A Vu, Trung-Nghia %A Dinh, H.Q. %A Hoang, Xuan-Huan %A VNU-UET, %B 2012 IEEE RIVF International Conference on Computing and Communication Technologies, Research, Innovation, and Vision for the Future (RIVF) %C Ho Chi Minh city %D 2012 %F SisLab:113 %K Accuracy;Biological cells;Genetic algorithms;Kernel;Optimization;Support vector machines;biology computing;genetic algorithms;support vector machines;Drosophila embryonic development;GA-SVM;biological experimental data;gene regulation;gene regulatory activity prediction improvement;genetic algorithm;parameter selection;sequencing technologies;support vector machine;transcriptional factor binding profiles; %P 1-4 %T GASVM: A Genetic Algorithm for Improving Gene Regulatory Activity Prediction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/113/ %X Gene regulatory activity prediction problem is one of the important steps to understand the significant factors for gene regulation in biology. The advents of recent sequencing technologies allow us to deal with this task efficiently. Amongst these, Support Vector Machine (SVM) has been applied successfully up to more than 80 accuracy in the case of predicting gene regulatory activity in Drosophila embryonic development. In this paper, we introduce a metaheuristic based on genetic algorithm (GA) to select the best parameters for regulatory prediction from transcriptional factor binding profiles. Our approach helps to improve more than 10 accuracy compared to the traditional grid search. The improvements are also significantly supported by biological experimental data. Thus, the proposed method helps boosting not only the prediction performance but also the potentially biological insights. %0 Journal Article %@ 2588-1086 %A Duong, Duc Tran %A Pham, Son Bao %A Tan, Hanh %D 2017 %F SisLab:2762 %I VNU %J VNU Journal of Science: Computer Science and Communication Engineering %N 1 %T Author Profiling of Vietnamese Forum Posts - An Investigation on Content-based Features %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2762/ %V 33 %X In this paper, we investigate the author profiling task for Vietnamese forum posts to predict demographic attributes, such as gender, age, occupation, and location of the author. Although we conducted the experiments on different types of features, including style-based and content-based features, we focus more on analysis the effects of content-based features. We used machine learning approaches to perform classification tasks on datasets we collected from popular forums in Vietnamese. The results show that these kinds of features work well on such a kind of short and free style messages as forum posts, in which, content-based features achieved much better results than style-based features. %0 Report %9 Technical Report %@ TR2017-FIT-003 %A Duong, Le Minh %A Tran, Van Truong %D 2017 %F SisLab:2896 %T Cải tiến tính năng multi-homming và congestion control cho giao thức SCTP trong mạng không dây đa bước %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2896/ %X Hiệu suất của các giao thức tầng giao vận trong mạng không dây đa bước (Multihop Wirless Network - MHWN) dựa trên chuẩn IEEE 802.11 nhận được nhiều sự quan tâm của các nhà nghiên cứu. Rất nhiều công trình nghiên cứu đã chỉ ra rằng các giao thức này có hiệu suất không cao trong MHWN, đặc biệt là TCP. Ra đời sau các giao thức được sử dụng cho mạng có dây như TCP, UDP, giao thức SCTP đã kế thừa được những ưu điểm và khắc phục một số nhược điểm của các giao thức truyền thống trên. Một số tính năng điển hình của SCTP có thể kể đến là multi-homing và congestion control. Đây là các tính năng được trang bị nhằm nâng cao hiệu năng hoạt động của giao thức. Trong bài báo này, chúng tôi tập trung vào đề xuất giải pháp cải tiến cho tính năng multi-homing và congestion control theo cách tiếp cận liên tầng nhằm nâng cao hiệu quả hoạt động của SCTP trong mạng MHWN. Thử nghiệm mô phỏng cho thấy SCTP với tính năng cải tiến hoạt động hiệu quả hơn giao thức nguyên bản. %0 Conference Paper %A Duong, Le Minh %A Veque, Veronique %A Zitoune, L. %B CFIP 2011 - Colloque Francophone sur l Ingénierie des Protocoles (2011) %C France %D 2011 %F SisLab:2278 %T Cross layer metrics for improving transport protocols in multihop wireless networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2278/ %X Our work concentrates on improving the performance of transport protocols in multihop wireless networks. We present different metrics from link and physic layers to improve the performance of TCP congestion control. This paper introduces a classification of cross layer metrics used to improve the transport level. %0 Conference Paper %A Duong, Le Minh %A Veque, Veronique %A Zitoune, L. %B 2012 IEEE 23rd International Symposium on Personal, Indoor and Mobile Radio Communications - (PIMRC) %D 2012 %F SisLab:2280 %K access protocols;multimedia communication;telecommunication congestion control;transport protocols;wireless LAN;wireless channels;IEEE 802.11 multihop wireless networks;MAC metric;MAC-aware rate control;MHWN;collision;congestion level;multimedia streaming applications;network load;packet loss;random channel errors;rate control transport protocols;route failures;transport layer performance;wireless medium characteristics;Delay;IEEE 802.11 Standards;Receivers;Throughput;Topology;Transport protocols %P 1436-1441 %T MAC-aware rate control for transport protocol in multihop wireless networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2280/ %X Transport layer performance in IEEE 802.11 multihop wireless networks (MHWNs) has been greatly challenged by wireless medium characteristics and multihop nature which are the sources of several types of packet loss including collision, random channel errors and route failures. Rate control transport protocols, the candidates for multimedia streaming applications suffer from high loss rates and end-to-end delay in MHWNs. A common research direction is that the rate control mechanisms at transport layer should be aware of MAC layer contention to keep the network load at a reasonable ;evel. In this paper, we introduce a new MAC metric which reflects the contention and congestion levels more accurately. The metric is then used to improve the rate control mechanism of a rate-based transport protocol in MHWNs. The simulation results show that the adapted mechanism introduces significant performance improvement in MHWNs. %0 Conference Paper %A Duong, Le Minh %A Veque, Veronique %A Zitoune, L. %B Wireless Communications and Mobile Computing Conference (IWCMC) %D 2012 %F SisLab:2279 %T A Medium Access Delay MAC aware Metric for Multihop Wireless Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2279/ %X Internet predominant transport protocols, such as TCP and TFRC, face performance degradation in Multihop Wireless Networks (MHWNs) because of the high loss and link failure rates. Many solutions have been proposed to improve the transport layer operation. These solutions are either based on network state estimation or use information from MAC layer (called MAC metrics) in a cross-layer manner to better comprehend the network state. In this paper, we define a new MAC metric called Medium Access Delay (MAD) to better reflect the network state, and provide a comparative study of MAD over other pertinent MAC metrics, their expected usage and measurement methods at MAC layer. We also investigate the behaviors of MAC metrics through several experiments in order to reveal their effectiveness in reflecting network events such as contention, collision and loss. %0 Conference Paper %A Duong, Le Minh %A Zitoune, L. %A Veque, Veronique %B 2013 IEEE 78th Vehicular Technology Conference (VTC Fall) %D 2013 %F SisLab:2281 %K protocols;telecommunication control;telecommunication standards;wireless LAN;BMRC;IEEE 802.11;MHWN;average transmission time;bimetric rate control;effective packet sending rate estimation;medium access delay;multihop wireless networks;network contention level;packet loss rate;random channel errors;rate-based protocols;route failures;transport layer performance;Delays;Network topology;Packet loss;Throughput;Topology %P 1-6 %T Improvement for Rate-Based Protocols in Multihop Wireless Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2281/ %X Transport layer performance in IEEE 802.11 multihop wireless networks (MHWNs) has been greatly challenged by wireless medium characteristics and multihop nature which induce several types of packet loss including collision, random channel errors and route failures. In this paper, we propose a novel rate control scheme, called Bi-Metric Rate Control (BMRC), which regulates efficiently the source rate in MHWNs. BMRC’s design is based on two MAC metrics: the Medium Access Delay used to detect the network contention level, and the Average Transmission Time used to estimate the effective packet sending rate by which the network will not be overloaded. The simulation results show that the adapted mechanism introduces significant performance improvement in terms of fairness, packet loss rate and delay in MHWNs. %0 Journal Article %@ 2156-5570 %A Duong, Ngoc Son %A Du, Phuong Hanh %A Nguyen, Ngoc Cuong %A Nguyen, Ngoc Hoa %D 2021 %F SisLab:4530 %I Science and Information (SAI) %J International Journal of Advanced Computer Science and Applications (IJACSA) %T A Graph Reduction Method for Improving the Computation of Betweenness Centrality in Large-Scale Graphs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4530/ %0 Book Section %A Duong, Tran Duc %A Pham, Bao Son %A Tan, Hanh %B Recent Developments in Intelligent Information and Database Systems %D 2016 %F SisLab:2257 %I Springer International Publishing %P 287-296 %T Using Content-Based Features for Author Profiling of Vietnamese Forum Posts %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2257/ %V 642 %X This paper reports the results of author profiling task for Vietnamese forum posts to identify the personal traits, such as gender, age, occupation, and location of the author using content-based features. Experiments were conducted on the different types of features, including stylometric features (such as lexical, syntactic, structural features) as well as content-based features (the most important words) to compare the performance and on the data sets we collected from the various forums in Vietnamese. Three learning methods, consisting of Decision Tree, Bayes Network, Support Vector Machine (SVM), were tested and the SVM achieved the best results. The results show that these kinds of features work well on such a kind of short and free style messages as forum posts, in which, content-based features yielded much better results than stylometric features. %0 Conference Paper %A Duong, Tuan Anh %A Sam, Duc Vu %A Pham, Ngoc Hung %B SSC: the 2016 Student Scientific Contest %C Hanoi %D 2016 %F SisLab:1498 %T Phương pháp và công cụ kiểm thử tự động cho các ứng dụng C/C++ %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1498/ %0 Conference Paper %A Duong, Viet Huy %A Nguyen, Dinh Viet %B IEEE International Conference on System Science and Engineering, ICSSE2017 %C Ho Chi Minh City, Vietnam %D 2017 %F SisLab:2681 %P 526-531 %T ATTS-DF: Adaptive Tracking Solution to the Target for Data Fusion in Wireless Sensor Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2681/ %X Wireless sensor networks are composed of many sensor nodes limited to residual energy. The sensor nodes are randomly scattered within in the ranger monitoring and sending - receiving data by radio waves. The measured data often contains the same information, and sending to base station causes the waste of energy of sensor nodes and the risk of congestion. According to the statistics, the energy consumed by transceiver radio signals is many times greater than the energy consumes to process other tasks of sensor node including the calculation on the sensor node. In this paper, we propose algorithms installed on the sensor nodes to monitor non-cyclical target fixed, which base on the fluctuation of the target when it exceeds the threshold. Moreover, algorithm control sensor nodes measures just in time fluctuations exceeded targets need to measure. Keywords— adaptive tracking, data fusion, ATTS-DF, WSNs %0 Conference Paper %A Duong, Viet Huy %A Nguyen, Dinh Viet %B KSE: the 2015 International Conference on Knowledge and Systems Engineering %C Ho Chi Minh city, Vietnam %D 2015 %F SisLab:1612 %P 1-6 %T DF-AMS: Proposed solutions for multi-sensor data fusion in wireless sensor networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1612/ %X When using multiple sensor nodes in wireless sensor networks (WSNs) for monitoring (measuring) parameters of the target and sending the result to the base station (BS), data redundancy is an inevitable problem. The measured data often contains the same information, and sending redundant data to BS causes the waste of energy of sensor nodes and the risk of congestion. Multi-sensor data fusion in WSNs is a technology of gathering and processing data applied from node to BS. It improves the performance of surveillance systems by allowing the obtained sensed information from multiple sensor nodes aggregated to one unified format data packet to send to BS to make decision. In this paper, we propose a solution namely DF-AMS (Data Fusion - Average Median Sampling) for sampling and fusing data from the sensor nodes in the cluster to optimize the energy conservation of sensor nodes in clusters and cluster head node. %0 Conference Paper %A Duong, Viet Huy %A Nguyen, Dinh Viet %B The 9th International Conference, Knowledge and Systems Engineering (KSE) %C Hue, Vietnam %D 2017 %F SisLab:2682 %P 62-67 %T DF-SWin: Sliding Windows for Multi-Sensor Data Fusion in Wireless Sensor Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2682/ %X Abstract— When using multiple sensor nodes in wireless sensor networks (WSNs) for monitoring (or measuring) parameters of the target and sending the result to the base station (BS), data redundancy is an inevitable problem. The measured data often contains the same information, and sending redundant data to BS causes the waste of energy of sensor nodes and the risk of congestion. Multi-sensor data fusion in WSNs is a technology of gathering and processing data applied from node to BS. It improves the performance of surveillance systems by allowing the obtained sensed information from multiple sensor nodes aggregated to one unified format data packet to send to BS to make decision. In this paper, we propose a solution namely DF-SWin for sampling sensor node in the cluster to optimize the energy conservation of sensor nodes in clusters and cluster head. Keywords— DF-SWin, sliding window, data fusion, WSNs %0 Journal Article %@ 0866-8612 %A Duong, Viet Huy %A Nguyen, Dinh Viet %D 2016 %F SisLab:2224 %I VNU Journal of Science %J VNU Journal of Computer Science and Communication Engineering %N 2 %P 63-71 %T Target tracking solution for multi-sensor data fusion in wireless sensor networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2224/ %V 32 %X Wireless sensor networks are often composed of many sensor nodes, they are powered by batteries with limited capacity . The sensor nodes are randomly scattered within in the ranger monitoring and send – receiver data using radio waves. Many research projects had demonstrated the consumption of battery power by the data transceiver occupy large compared to the calculation on the sensor node. In this paper, we propose energy saving solutions of nodes in a cluster by only choosing some nodes in the cluster to track the target and transmit this data to cluster head nodes. We based on the location of the sensor nodes in the cluster compare with the location of target and cluster head nodes to perform this selection. The effectiveness of the proposed solutions will be evaluated based on the number of sensor nodes are selected considering the number of nodes in the cluster, this is the base for the effectiveness of energy saving as well as the cluster nodes. Keywords: Target tracking; multi-sensor data fusion, WSNs, ETR-DF. %0 Conference Paper %A Duong, Viet Huy %A Nguyen, Dinh Viet %B the 8th National Conference on Fundamental and Applied IT Research %D 2015 %F SisLab:1613 %T Đề xuất giải pháp tiền xử lý để tổng hợp dữ liệu nhiều cảm biến trong mạng cảm biến không dây %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1613/ %0 Conference Paper %A Duong, Viet Huy %A Nguyen, Dinh Viet %B Hội thảo Quốc gia lần thứ XVII Một số vấn đề chọn lọc của Công nghệ thông tin và Truyền thông %C Việt Nam %D 2014 %F SisLab:2982 %P 50-55 %T Đề xuất giải pháp tổng hợp dữ liệu nhiều cảm biến trong mạng cảm biến không dây. %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2982/ %X Hiện nay, đối với mạng cảm biến không dây (Wireless Sensor Networks) kỹ thuật phân cụm và sử dụng nút chủ cụm (cluster head) để tham gia định tuyến được xem là một trong những giải pháp tối ưu năng lượng cho mạng. Tuy nhiên, kết hợp phân nhóm trong định tuyến với giải pháp tổng hợp dữ liệu (data fusion hoặc data aggregation) nhiều nút cảm biến trong nhóm đó có thể đạt hiệu quả tốt hơn nữa về tiết kiệm năng lượng. Trong bài báo này, chúng tôi đề xuất giải pháp DF-Tab để tổng hợp dữ liệu nhiều cảm biến trong mạng cảm biến không dây có phân cụm và mô phỏng bằng NS2 giải thuật DF-Tab với cấu hình mạng cảm biến tương tự giải thuật LEACH. Theo kết quả mô phỏng, so với kỹ thuật LEACH, giải pháp DF-Tab đã tiết kiệm được 25%-40% năng lượng của nút CH. %0 Conference Paper %A Duong, Viet Huy %A Nguyen, Duy Tan %A Ho, Duc Ai %A Nguyen, Dinh Viet %B Hội nghị Khoa học Công nghệ Quốc gia lần thứ VII về Nghiên cứu cơ bản và ứng dụng Công nghệ thông tin (FAIR) %D 2014 %F SisLab:2237 %P 668-677 %T Tiếp cận phương pháp tổng hợp dữ liệu nhiều cảm biến trong mạng cảm biến không dây bằng lý thuyết tập thô %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2237/ %V 4 %X Mạng cảm biến không dây bao gồm nhiều nút cảm biến sử dụng đường truyền không dây, thường được rải ngẫu nhiên trong vùng cần giám sát, thời gian hoạt động của mỗi nút phụ thuộc vào dung lượng nguồn pin dự trữ. Dữ liệu mà các nút cảm biến thu thập được cần được truyền đến một nút đích; có thể có nhiều nút cảm biến cùng gửi về nút đích dữ liệu liên quan đến một tham số của đối tượng cần giám sát. Để trạm đích nhận được thông tin phản ánh đúng bản chất sự kiện xảy ra từ các nút cảm biến đồng thời hạn chế tối đa các dữ liệu dư thừa (do cùng thông tin) truyền đi trong mạng gây tiêu tốn năng lượng vô ích và nguy cơ nghẽn mạng, tổng hợp dữ liệu là một trong những lựa chọn để giải quyết vấn đề đó. Trong bài báo này, chúng tôi đề xuất quy trình và mô hình ứng dụng l ý thuyết tập thô để tổng hợp dữ liệu nhiều cảm biến trong mạng cảm biến không dây. %0 Conference Paper %A Fernandez, Facundo %A Gaul, David %A Jones, Christina %A Monge, Maria Eugenia %A Paine, Martin %A Tran, Quoc Long %A McDonald, John %B ABSTRACTS OF PAPERS OF THE AMERICAN CHEMICAL SOCIETY %D 2015 %F SisLab:3278 %T Phenotyping of early stage ovarian cancer by mass spectrometry imaging and untargeted metabolomics %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3278/ %V 250 %0 Journal Article %@ 2045-2322 %A Gaul, David %A Mezencev, Roman %A Tran, Quoc Long %A Jones, Christina M. %A Benigno, Benedict %A Gray, Alexander %A Fernandez, Facundo %A McDonald, John F. %D 2015 %F SisLab:1469 %I Nature Publishing Group %J Scientific Reports %N 16351 %T Highly-Accurate Metabolomic Detection of Early-Stage Ovarian Cancer %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1469/ %V 5 %0 Conference Paper %A Giang, Thanh Trung %A Tran, Dang Hung %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1521 %T Tích hợp dữ liệu sử dụng học đa nhân kết hợp với giảm chiều dữ liệu thực hiện phân cụm các bệnh nhân ung thư. %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1521/ %X Mặc dù các nghiên cứu về ung thư đang diễn ra nhưng các phương pháp điều trị hiện có là rất hạn chế về số lượng và hiệu quả, ngoài ra quyết định điều trị cho các bệnh nhân vẫn là một vấn đề khó khăn. Việc thành lập các phân nhóm giúp trợ giúp ra quyết định chủ yếu vẫn dựa trên các loại dữ liệu riêng biệt. Tuy nhiên, việc phân tích dữ liệu bệnh nhân đa chiều dựa trên nhiều số đo đặc trưng phân tử khác nhau như: biểu thức Gene, DNA Methylation, biểu thức miRNA có thể giúp phát hiện các đặc điểm nội tại của các khối u chính xác hơn. Một phương pháp giúp tích hợp dữ liệu từ nhiều dạng đặc tả khác nhau dựa trên phương pháp học đa nhân đồng thời kết hợp với việc giảm chiều dữ liệu được chúng tôi sử dụng trong báo cáo này nhằm tạo nên một bộ dữ liệu có tính tổng quát cao hơn từ nhiều loại biểu diễn dữ liệu khác nhau. Sau khi tích hợp dữ liệu, chúng tôi tiến hành phân cụm bệnh nhân ung thư dựa trên dữ liệu đã tích hợp và đánh giá kết quả phân cụm thu được. Ngoài ra chúng tôi cũng đưa ra các gợi ý về mặt lựa chọn hàm nhân cho các loại dữ liệu cũng như các tham số để đạt được kết quả phân cụm tốt nhất cho mô hình này. %0 Conference Paper %A Gogolla, Martin %A Dang, Duc Hanh %A Fabian, Buttner %B The 3rd International Conference on Applications of Graph Transformations with Industrial Relevance (AGTIVE) %D 2007 %F SisLab:1967 %I Springer %P 585-586 %T From Graph Transformation to OCL Using USE %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1967/ %V 5088 %0 Journal Article %@ 0305-0548 %A Ha, Minh Hoang %A Nguyen, Tat Dat %A Nguyen, Duy Thinh %A Pham, Hoang Giang %A Do, Thuy %A Rousseau, Louis-Martin %D 2020 %F SisLab:4061 %I Elsevier %J Computers & Operations Research %T A new constraint programming model and a linear programming-based adaptive large neighborhood search for the vehicle routing problem with synchronization constraints %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4061/ %V 124 %X We consider a vehicle routing problem which seeks to minimize cost subject to time window and synchronization constraints. In this problem, the fleet of vehicles is categorized into regular and special vehicles. Some customers require both vehicles’ services, whose service start times at the customer are synchronized. Despite its important real-world application, this problem has rarely been studied in the literature. To solve the problem, we propose a Constraint Programming (CP) model and an Adaptive Large Neighborhood Search (ALNS) in which the design of insertion operators is based on solving linear programming (LP) models to check the insertion feasibility. A number of acceleration techniques is also proposed to significantly reduce the computational time. The computational experiments show that our new CP model finds better solutions than an existing CP-based ALNS, when used on small instances with 25 customers and with a much shorter running time. Our LP-based ALNS dominates the CP-based ALNS, in terms of solution quality, when it provides solutions with better objective values, on average, for all instance classes. This demonstrates the advantage of using linear programming instead of constraint programming when dealing with a variant of vehicle routing problems with relatively tight constraints, which is often considered to be more favorable for CP-based methods. We also adapt our algorithm to solve a well-studied variant of the problem, and the obtained results show that the algorithm provides good solutions as state-of-the-art approaches and improves four best known solutions. %0 Conference Paper %A Ha, Minh Hoang %A Vu, Duy Manh %A Zinder, Yakov %A Nguyen, Trung Thanh %B KSE 2019 %C Da Nang %D 2019 %F SisLab:3674 %T On The Capacitated Scheduling Problem with Conflict Jobs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3674/ %X The paper is concerned with scheduling jobs on parallel identical machines under the restrictions imposed by a conflict graph. The nodes of this conflict graph represent jobs and each edge indicates that the jobs associated with the nodes, incident to this edge, can not be processed concurrently. The jobs have a common due date and each job has the associated weight. The goal is to maximise the total weight of jobs which completion times do not exceed the due date. The considered scheduling model was motivated by the problem arising in the telecommunication industry. The paper identifies polynomially solvable and NP-hard particular cases and presents two mixed integer linear programming formulations together with their comparison by means of computational experiments. %0 Journal Article %A Ha, Quang Minh %A Deville, Yves %A Pham, Quang Dung %A Ha, Minh Hoang %D 2019 %F SisLab:3671 %I Springer %J Journal of Heuristics %T A hybrid genetic algorithm for the traveling salesman problem with drone %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3671/ %X This paper addresses the traveling salesman problem with drone (TSP-D), in which a truck and drone are used to deliver parcels to customers. The objective of this problem is to either minimize the total operational cost (min-cost TSP-D) or minimize the completion time for the truck and drone (min-time TSP-D). This problem has gained a lot of attention in the last few years reflecting the recent trends in a new delivery method among logistics companies. To solve the TSP-D, we propose a hybrid genetic search with dynamic population management and adaptive diversity control based on a split algorithm, problem-tailored crossover and local search operators, a new restore method to advance the convergence and an adaptive penalization mechanism to dynamically balance the search between feasible/infeasible solutions. The computational results show that the proposed algorithm outperforms two existing methods in terms of solution quality and improves many best known solutions found in the literature. Moreover, various analyses on the impacts of crossover choice and heuristic components have been conducted to investigate their sensitivity to the performance of our method. %0 Conference Paper %A Ha, Quang Thuy %A Bui, Hong Nhung %A Nguyen, Tri Thanh %B The 8th International Conference on Computational Collective Intelligence (ICCCI) %C Halkidiki, Greece %D 2016 %F SisLab:1868 %T A trace clustering solution based on using the distance graph model %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1868/ %0 Conference Paper %A Ha, Quang Thuy %A Ha, Thi Oanh %A Nguyen, Thi Dung %A Nguyen, Thuy Linh %A VNU-UET, %B The 4th International Conference on Computational Collective Intelligence (ICCCI 2012) %C Ho Chi Minh, Vietnam %D 2012 %F SisLab:79 %P 335-344 %T Refining the Judgement Threshold to Improve Recognizing Textual Entailment Using Similarity %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/79/ %0 Conference Paper %A Ha, Quang Thuy %A Hoang, Thi Lan Giao %A Nguyen, Linh Anh %A Nguyen, Hung Son %A Szalas, Andrzej %A Tran, Thanh-Luong %B 2012: the 3rd International Symposium on Information and Communication Technology (SoIC) %C Ha Long, Vietnam %D 2012 %F SisLab:82 %P 241-249 %T A Bisimulation-based Method of Concept Learning for Knowledge Bases in Description Logics %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/82/ %0 Book Section %A Ha, Quang Thuy %A Nguyen, Linh Anh %A Nguyen, Thi Hong Khanh %A Tran, Thanh Luong %A Faculty of Information TechnologyVNU University of Engineering and TechnologyHanoiVietnam, %A Division of Knowledge and System Engineering for ICT, Faculty of Information TechnologyTon Duc Thang UniversityHo Chi Minh CityVietnam, %A Institute of InformaticsUniversity of WarsawWarsawPoland, %A Faculty of Information TechnologyElectricity Power UniversityHanoiVietnam, %A Department of Information TechnologyUniversity of Sciences, Hue UniversityHueVietnam, %B International Joint Conference on Rough Sets (IJCRS 2018) %D 2018 %E Nguyen, Hung Son %E Ha, Quang-Thuy %E Li, Tianrui %E Malgorzata, Przybyla-Kasperek %F SisLab:3156 %I Springer %N 11103 %P 559-571 %S Lecture Notes in Computer Science %T Fuzzy Bisimulations in Fuzzy Description Logics Under the Gödel Semantics %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3156/ %X Description logics (DLs) are a suitable formalism for representing knowledge about domains in which objects are described not only by attributes but also by binary relations between objects. Fuzzy DLs can be used for such domains when data and knowledge about them are vague. One of the possible ways to specify classes of objects in such domains is to use concepts in fuzzy DLs. As DLs are variants of modal logics, indiscernibility in DLs is characterized by bisimilarity. The bisimilarity relation of an interpretation is the largest auto-bisimulation of that interpretation. In (fuzzy) DLs, it can be used for concept learning. In this paper, for the first time, we define fuzzy bisimulation and (crisp) bisimilarity for fuzzy DLs under the Gödel semantics. The considered logics are fuzzy extensions of the DL ALCreg with additional features among inverse roles, nominals, qualified number restrictions, the universal role and local reflexivity of a role. We give results on invariance of concepts as well as conditional invariance of TBoxes and ABoxes for bisimilarity in fuzzy DLs under the Gödel semantics. We also provide a theorem on the Hennessy-Milner property for fuzzy bisimulations in fuzzy DLs under the Gödel semantics. %0 Journal Article %@ 1859-4794 %A Ha, Quang Thuy %A Nguyen, Thanh Thuy %A Pham, Bao Son %A Phan, Xuan Hieu %A Tran, Trong Hieu %A Tran, Mai Vu %A Tran, Quoc Long %A Nguyen, Tri Thanh %A Ly, Hoang Tung %A Trường Đại học Công nghệ, ĐHQGHN, %A Bộ Khoa học và Công nghệ, %D 2020 %F SisLab:4038 %I Bộ Khoa học và Công nghệ %J Tạp chí Khoa học và Công nghệ Việt Nam %T Môi trường và giải pháp cho chiến lược quốc gia về trí tuệ nhân tạo của Việt Nam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4038/ %X Trí tuệ nhân tạo (TTNT) đã tạo ra những chuyển đổi lớn về kinh tế, xã hội, đời sống của con người, và nhiều nước đã xây dựng chiến lược quốc gia về TTNT. Phân tích môi trường và xây dựng chiến lược quốc gia về TTNT là một công việc đầy thách thức đối với nhiều quốc gia, trong đó có Việt Nam. Bài báo tìm hiểu một số nghiên cứu phân tích về môi trường và xây dựng chiến lược quốc gia về TTNT. Từ đó đưa ra giải pháp định hướng cho chiến lược quốc gia về TTNT của Việt Nam %0 Journal Article %@ 0866-7756 %A Ha, Quang Thuy %A Nguyen, Tri Thanh %D 2018 %F SisLab:2956 %J Tạp chí Công Thương - Các kết quả nghiên cứu khoa học và Ứng dụng công nghệ %P 160-165 %T Khoa học dữ liệu, tiền hóa dữ liệu và khả năng thực thi tại Việt Nam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2956/ %V 3 %0 Journal Article %@ 2615-9759 %A Ha, Quang Thuy %A Pham, Bao Son %A Phan, Xuan Hieu %A Tran, Trong Hieu %A Tran, Mai Vu %A Nguyen, Tri Thanh %A Pham, Van Dung %A Tran, The Hiep %A Trường Đại học Công nghệ, ĐHQGHN, %A Trường Đại học Kinh tế, ĐHQGHN, %D 2020 %F SisLab:4084 %I Bộ Khoa học và Công nghệ %J Khoa học và Công nghệ Việt Nam %T Những yếu tố quan trọng trong Chiến lược quốc gia về trí tuệ nhân tạo %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4084/ %X Chuyển đổi số và trí tuệ nhân tạo (TTNT) là các quá trình tất yếu, song cũng rất phức tạp đối với các tổ chức và các quốc gia. Nhóm tác giả đã chỉ ra 3 yếu tố (trưởng thành số, kiến trúc tổ chức chuyển đổi cân đối - CĐCĐ, tài năng con người về TTNT) đóng vai trò quan trọng đặc biệt trong chiến lược quốc gia về TTNT, trong chiến lược và sáng kiến chuyển đổi số và TTNT của tổ chức. %0 Conference Paper %A Ha, Quang Thuy %A Pham, Thi Ngan %A Nguyen, Van Quang %A Nguyen, Minh Chau %A Pham, Thanh Huyen %A Nguyen, Tri Thanh %B 10th International Conference on Computational Collective Intelligence (ICCCI 2018) %C Bristol, United Kingdom %D 2018 %F SisLab:2960 %T o A new text semi-supervised multi-label learning model based on using the label-feature relations %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2960/ %X Multi-label learning has become popular and omnipresent in many real-world problems, especially in text classification applications, in which an instance could belong to different classes simultaneously. Due to these label constraints, there are some challenges occurring in building multi-label data. Semi-supervised learning is one possible approach to exploit abundantly unlabeled data for enhancing the classification performance with a small labeled dataset. In this paper, we propose a solution to select the most influential label based on using the relations among the labels and features to a semi-supervised multi-label classification algorithm on texts. Experiments on two datasets of Vietnamese reviews and English emails of Enron show the positive effects of the proposal. %0 Conference Paper %A Ha, Quang Thuy %A Pham, Thi Ngan %A Nguyen, Van Quang %A Vuong, Thi Hong %A Tran, Minh Tuoi %A Nguyen, Thi Cham %A Nguyen, Tri Thanh %B 10th Asian Conference on Intelligent Information and Database Systems (ACIIDS) %D 2018 %F SisLab:2750 %T A New Lifelong Topic Modeling Method and Its Application to Vietnamese Text Multi-label Classification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2750/ %X Lifelong machine learning is emerging in recent years thanks to its ability to use past knowledge for current problem. Lifelong topic modeling algorithms, such as LTM and AMC, are proposed and they are very useful. However, these algorithms focus on learning bias on the topic level not the domain level. This paper proposes a lifelong topic modeling method, which focuses on learning bias on the domain level based on a proposed domain closeness measure, and an application framework for multi-label classification on Vietnamese texts. Experimental results on three previously solved Vietnamese texts, and five different current Vietnamese text datasets in combination with different topic set sizes showed that our proposed method is better than AMC method for all cases %0 Journal Article %@ 1859-4794 %A Ha, Quang Thuy %A Phan, Xuan Hieu %A Nguyen, Tri Thanh %A Tran, Trong Hieu %A Tran, Mai Vu %A Lê, Đức-Trọng %A VNU-UET, %D 2021 %F SisLab:4489 %I Bộ Khoa học và Công nghệ %J Tạp chí Khoa học và Công nghệ Việt Nam %T Mô hình quản trị đại học châu Âu và liên hệ với Việt Nam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4489/ %X Qua khảo sát mô hình quản trị trường đại học (QTĐH) thuộc 29 hệ thống giáo dục đại học châu Âu, các tác giả đã liên hệ với Việt Nam và cho thấy, các nhà cải cách giáo dục Việt Nam chưa đặt cải cách vào bối cảnh phức tạp với rất nhiều tác động và mối quan hệ của hàng loạt yếu tố đa dạng liên quan đến dạy và học, đặc biệt, đã bỏ qua mối quan hệ phụ thuộc biện chứng của ba cột trụ là nhà hoạch định chính sách, giảng viên và sinh viên. Từ đó, các tác giả đã đưa ra một số khuyến nghị nhằm thúc đẩy sự phát triển hệ thống đại học của Việt Nam. %0 Journal Article %@ 0866-7756 %A Ha, Quang Thuy %A Phan, Xuan Hieu %A Nguyen, Tri Thanh %A Tran, Trong Hieu %A Tran, Mai Vu %A Nguyen, Huu Duc %A Trường Đại học Công nghệ, ĐHQGHN, %D 2020 %F SisLab:4039 %I Bộ Công thương %J Tạp chí Công Thương - Các kết quả nghiên cứu khoa học và Ứng dụng công nghệ %T Kinh tế số: Bối cảnh thế giới và liên hệ với Việt Nam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4039/ %X Sự phát triển mạnh mẽ song hành giữa các công nghệ số hóa (điển hình là Internet vạn vật) và kinh tế số (điển hình là sự xuất hiện và phát triển của nhiều hình thức kinh doanh sáng tạo và đa dạng) đã tạo nên sự biến đổi to lớn và nhanh chóng đối với kinh tế, xã hội và mọi mặt của đời sống con người. Kinh tế số đã trở thành một thành phần kinh tế đóng góp một tỷ trọng đáng kể với xu thế ngày càng tăng trong toàn bộ nền kinh tế quốc dân. Đồng thời, kinh tế số cũng làm phát sinh không ít các thách thức đáng quan tâm về khái niệm kinh tế số, mô hình phát triển kinh tế số và đo lường kinh tế số. Việc tìm ra các mô hình (khái niệm, phát triển, đo lường) kinh tế số nhận được sự quan tâm đặc biệt của nhiều quốc gia và các tổ chức quốc tế, như: Tổ chức hợp tác và phát triển kinh tế thế giới, Ngân hàng thế giới,... Bài báo này cung cấp một số tìm hiểu bước đầu về kinh tế số, về một số cơ hội và thách thức từ kinh tế số, cũng như mô hình đo lường. Bài báo cũng đưa ra một vài trao đổi sơ bộ về kinh tế số tại Việt Nam, đó là: tăng cường nhận thức đúng đắn và toàn diện về kinh tế số, phát triển nguồn nhân lực chuyên nghiệp kinh tế số (đặc biệt là lực lượng chuyên gia cao cấp về kinh tế số) và hệ sinh thái kinh tế số tại Việt Nam sẽ góp phần đẩy nhanh phát triển kinh tế số của Việt Nam. %0 Conference Paper %A Ha, Quang Thuy %A Vo, Dinh Hieu %A Phan, Xuan Hieu %A Nguyen, Tri Thanh %A Nguyen, Huu Duc %A Trường Đại học Công nghệ, ĐHQGHN, %B Phát triển mô hình trung tâm tri thức số cho các thư viện Việt Nam %C Đại học Quốc gia Hà Nội %D 2020 %F SisLab:4173 %P 132-145 %T Đo lường công bố khoa học: bối cảnh thế giới và liên hệ với Việt Nam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4173/ %X Đo lường công bố khoa học của cá nhân nhà khoa học, tổ chức khoa học là một chủ đề hấp dẫn theo các khía cạnh khoa học, kinh tế và quản lý. Tuy nhiên, đo lường công bố khoa học cũng gặp không ít thách thức, vì vậy, việc tìm hiểu một cách toàn diện về đo lường công bố khoa học là một công việc cần thiết. Bài viết này cung cấp một khảo sát bước đầu song đủ khái quát về đo lường công bố khoa học, một số thách thức chính và xu hướng hoạt động này trong thời đại số. Một số trao đổi về hoạt động đo lường công bố khoa học tại Việt Nam cũng được trình bày. %0 Journal Article %@ 1687-7268 %A Ha, Su Wook %A Lee, Yang Koo %A Vu, Thi Hong Nhan %A Jung, Young Jin %A Keun, Ho Ryu %D 2012 %F SisLab:2114 %I Hindawi %J Journal of Sensors %T An Environmental Monitoring System for Managing Spatiotemporal Sensor Data over Sensor Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2114/ %0 Conference Paper %A Ha, Su Wook %A Lee, Yang Koo %A Vu, Thi Hong Nhan %A Ryu, Keun Ho %B 6th International Conference on Computer Sciences and Convergence Information Technology, IEEE %C Korea %D 2011 %F SisLab:2130 %T Spatiotemporal Data Stream Management over Sensor Network Application %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2130/ %0 Conference Paper %A Ha, Van Sang %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1522 %T Phương pháp trích trọn đặc trưng hiệu quả cho dữ liệu có số chiều lớn %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1522/ %X Phân lớp là một trong những bài toán cơ bản trong khai phá tri thức và dữ liệu. Một thách thức của bài toán phân lớp là số lượng thuộc tính thường rất lớn, việc phân lớp sao cho chính xác và hiệu quả hiện vẫn là một nghiên cứu thú vị cho các nhà khoa học trong lĩnh vực khoa học máy tính. Bài báo đi sâu vào nghiên cứu giải thuật phân lớp random forest (RF). Đây là một giải thuật đã được nhiều nghiên cứu chứng minh là rất hiệu quả trong phân lớp đối với bộ dữ liệu có số lượng thuộc tính lớn. Trên cơ sở đó chúng tôi đề xuất một phương pháp học máy cho giải thuật phân lớp này nhằm tăng hiệu quả phân lớp của thuật toán. Cách tiếp cận này về cơ bản đã làm tăng khả năng phân lớp của giải thuật RF, phương pháp đề xuất còn cho thấy khả năng phân lớp tốt hơn một số phương pháp trích chọn đã được công bố. Như vậy, hướng cải tiến mà chúng tôi đề xuất là có khả thi và thu được kết quả tương đối cao. %0 Book Section %A Ha, Van Sang %A Nguyen, Ha Nam %B Machine Learning and Data Mining in Pattern Recognition %D 2016 %F SisLab:1941 %I Springer International Publishing %P 459-467 %T C-KPCA: Custom Kernel PCA for Cancer Classification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1941/ %X Abstract Principal component analysis (PCA) is an effective and well-known method for reducing high-dimensional data sets. Recently, KPCA (Kernel PCA), a nonlinear form of PCA, has been introduced into many fields. In this paper, we propose a new gene selection, namely Custom Kernel principal component analysis (C-KPCA). The new kernel function for KPCA is created by combining a set of kernel functions. First, Singular Value Decomposition (SVD) is used to reduce the dimension of microarray data. Input space is then mapped to ... %0 Conference Paper %A Ha, Van Sang %A Nguyen, Ha Nam %B International Conference on Nature of Computation and Communication %D 2016 %F SisLab:1939 %P 133-142 %T FRFE: Fast Recursive Feature Elimination for Credit Scoring %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1939/ %X Abstract Credit scoring is one of the most important issues in financial decision-making. The use of data mining techniques to build models for credit scoring has been a hot topic in recent years. Classification problems often have a large number of features, but not all of them are useful for classification. Irrelevant and redundant features in credit data may even reduce the classification accuracy. Feature selection is a process of selecting a subset of relevant features, which can decrease the dimensionality, reduce the running time, and ... %0 Journal Article %A Ha, Van Sang %A Nguyen, Ha Nam %A Nguyen, Duc Nhan %D 2016 %F SisLab:1942 %J Indian Journal of Science and Technology %N 20 %T A Novel Credit Scoring Prediction Model based on Feature Selection Approach and Parallel Random Forest %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1942/ %V 9 %X Background/Objectives: This article presents a method of feature selection to improve the accuracy and the computation speed of credit scoring models. Methods/Analysis: In this paper, we proposed a credit scoring model based on parallel Random Forest classifier and feature selection method to evaluate the credit risks of applicants. By integration of Random Forest into feature selection process, the importance of features can be accurately evaluated to remove irrelevant and redundant features. Findings: In this research, an algorithm to ... %0 Journal Article %@ 1947-9328 %A Hien, Dang Thi Thu %A Huan, Hoang Xuan %A Hoang, Le Xuan Minh %D 2015 %F SisLab:1634 %J International Journal of Operations Research and Information Systems %N 4 %P 57-74 %T An Effective Solution to Regression Problem by RBF Neuron Network %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1634/ %V 6 %X Radial Basis Function RBF neuron network is being applied widely in multivariate function regression. However, selection of neuron number for hidden layer and definition of suitable centre in order to produce a good regression network are still open problems which have been researched by many people. This article proposes to apply grid equally space nodes as the centre of hidden layer. Then, the authors use k-nearest neighbour method to define the value of regression function at the center and an interpolation RBF network training algorithm with equally spaced nodes to train the network. The experiments show the outstanding efficiency of regression function when the training data has Gauss white noise. %0 Conference Paper %A Ho, Huy Hung %A Nguyen, Ngoc Sinh %A Bui, Duy Hieu %A Tran, Xuan Tu %B The 4th NAFOSTED Conference on Information and Computer Science (NICS) %C Hanoi, Vietnam %D 2017 %F SisLab:2746 %P 201-206 %T Accurate and Low Complex Cell Histogram Generation by Bypass the Gradient of Pixel Computation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2746/ %X Histogram of Oriented Gradient (HOG) is a popular feature description for the purpose of object detection. However, HOG algorithm requires a performance system because of its complex operation set. Especially, the cell histogram generation is one of the most complex part, it uses inverse tangent, square, square root, floating point multiplication. In this paper, we propose an accurate and low complex cell histogram generation by bypass the gradient of pixel computation. It employs the bin’s boundary angle method to determine the two quantized angles. However, instead of choosing an approximate value of tan, the nearest greater and the nearest smaller of each tan values from ratio between pixel’s derivative in y and x direction are used. Magnitude of two bins are solutions of a system of two equations, which represent the equality of the gradient of a pixel and its two bins in both vertical and horizontal direction. The proposed method spends only 30 addition and 40 shift operations to identify two bins of a pixel. Simulation results show that the percentage error when reconstructing the difference in x and y direction are always less than 2% with 8-bit length of the fractional part. Additionally, manipulating the precision of gradient magnitude is very simple by pre-defined sine and cosine values of quantized angles. Synthesizing the hardware implementation presents that its area cost is 3.57 KGates with 45nm NanGate standard cell library. The hardware module runs at the maximum frequency of 400 MHz, and the throughput is 0.4 (pixel/ns) for a single module. It is able to support about 48 fps with 4K UHD resolution. %0 Report %9 Technical Report %@ TR2017-FIT-001 %A Ho, Phuong %D 2017 %F SisLab:2800 %T Catalan Number %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2800/ %X Dãy số Catalan là một trong những dãy số quan trọng trong Tổ hợp. Bài viết này trình bày một số bài toán liên quan đến dãy số Catalan và cách chứng minh một số là số Catalan. Hệ thống những bài tập này được sử dụng trong việc huấn luyện các đội tuyển thi IOI, ACM. %0 Conference Paper %A Ho, Thi Nga %A Can, Duy Cat %A Chng, Eng Siong %B International Conference on Asian Language Processing (IALP 2018) %C Bandung, Indonesia %D 2018 %F SisLab:3170 %T An Investigation of Word Embeddings with Deep Bidirectional LSTM for Sentence Unit Detection in Automatic Speech Transcription %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3170/ %X This work investigates the effectiveness of using the word based and sub-word based embedding representations as input for a deep bidirectional Long Short-Term Memory Network for Sentence Unit Detection in Automatic Speech Recognition transcription. Our experimental results show that using sub-word based embedding can significantly improve the SUD performance when a limited text is used to train both the word embedding and the SUD model. The SUD model using the sub-word based embedding gains up to 2.07% absolute improvement in F1-score as compared to the best model trained with the word-based embedding. When tested on a domain-mismatch condition, the SUD model with sub-word based embedding trained from the in-domain data gives an approximate 2% and 1% improvement over the best model using out-of-domain embedding with reference and ASR transcription with 29.5% Word Error Rate respectively. %0 Conference Paper %A Hoa, Nong Thi %A Bui, The Duy %A Dang, Trung Kien %B 10th International Symposium on Neural Networks (ISNN 2013), LNCS Springer %D 2013 %F SisLab:784 %T Efficiency improvements for fuzzy associative memory %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/784/ %V 2951 %0 Conference Paper %A Hoang, Cuong %A Le, Anh Cuong %A Nguyen, Phuong Thai %A Ho, Tu Bao %A VNU-UET, %B 2012 IEEE RIVF International Conference on Computing and Communication Technologies, Research, Innovation, and Vision for the Future (RIVF) %C Ho Chi Minh city %D 2012 %F SisLab:114 %K Electronic publishing;Encyclopedias;Error analysis;Internet;Length measurement;Training;language translation;learning (artificial intelligence);nonparallel corpora;parallel sentence pair detection;reinforcement learning scheme;statistical machine translation system; %P 1-6 %T Exploiting Non-Parallel Corpora for Statistical Machine Translation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/114/ %X Constructing a corpus of parallel sentence pairs is an important work in building a Statistical Machine Translation system. It impacts deeply how the quality of a Statistical Machine Translation could achieve. The more parallel sentence pairs we use to train the system, the better translation's quality it is. Nowadays, comparable non-parallel corpora become important resources to alleviate scarcity of parallel corpora. The problem here is how to extract parallel sentence pairs automatically but accurately from comparable non-parallel corpora, which are usually very "noisy". This paper presents how we can apply the reinforcement-learning scheme with our new proposed algorithm for detecting parallel sentence pairs. We specify that from an initial set of parallel sentences in a domain, the proposed model can extract a large number of new parallel sentence pairs from non-parallel corpora resources in different domains, concurrently increasing the system's translation ability gradually. %0 Journal Article %A Hoang, Cuong %A Le, Anh Cuong %A Nguyen, Phuong Thai %A Pham, Bao Son %A Ho, Tu Bao %D 2014 %F SisLab:1180 %J Fundamenta Informaticae %P 179-199 %T An Efficient Framework for Extracting Parallel Sentences from Non-Parallel Corpora %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1180/ %V 130 %0 Conference Paper %A Hoang, Cuong %A Le, Anh Cuong %A Pham, Bao Son %A VNU-UET, %B 2012 Fourth International Conference on Knowledge and Systems Engineering (KSE) %C Danang, Vietnam %D 2012 %F SisLab:98 %K Analytical models;Educational institutions;Hidden Markov models;Mathematical model;Systematics;Training;computational linguistics;language translation;natural language processing;statistical analysis;word processing;heuristic method;lexical modelling;lexical translation;phrase learning;statistical English-Vietnamese phrase-based translation;statistical alignment model;statistical phrase-based machine translation;word alignment model;word-based translation model; %P 143 -150 %T A Systematic Comparison between Various Statistical Alignment Models for Statistical English-Vietnamese Phrase-Based Translation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/98/ %X In statistical phrase-based machine translation, the step of phrase learning heavily relies on word alignments. This paper provides a systematic comparison of applying various statistical alignment models for statistical English-Vietnamese phrase-based machine translation. We will also invest a heuristic method for elevating the translation quality of using higher word-alignment models by improving the quality of lexical modelling. In detail, we will experimentally show that taking up the lexical translation seems to be an appropriate approach to force "higher" word-based translation models be able to efficiently "boost" their merits. We hope this work will be a reliable comparison benchmark for other studies on using and improving the statistical alignment models for English-Vietnamese machine translation systems. %0 Journal Article %@ 2162-1039 %A Hoang, Dang Kien %A Nguyen, Dai Tho %A Nguyen, Thi Thu Trang %F SisLab:4690 %I IEEE %J 2021 International Conference on Advanced Technologies for Communications (ATC) %T Efficient Incremental Instance-based Learning Algorithms for Open World Malware Classification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4690/ %X Malware is growing rapidly in number and become more and more sophisticated. To prevent them we need to collect samples continuously and update them to the classifier. In this paper, we will propose a method to update new labeled samples of malware to the classifier easily without re-training everything. The classifier can be updated by both labeled malware samples of an existing class or a new class. Our method also has the ability to detect samples of unknown families. Experiments are performed over the traditional computer malware dataset and the IoT malware dataset. The results have shown that our method can reach the macro F1-score almost the same re-train everything but take significantly less time. %0 Conference Paper %A Hoang, Dang Kien %A Nguyen, Dai Tho %A Vu, Duy Loi %B The 2020 IEEE-RIVF International Conference on Computing and Communications Technologies %C Ho Chi Minh %D 2020 %F SisLab:3987 %T IoT Malware Classification Based on System Calls %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3987/ %X IoT devices play an important role in the industrial revolution 4.0. However, this type of device may exhibit specific security vulnerabilities that can be easily exploited to cause botnet attacks and other malicious activities. In this paper, we introduce a new method for classification and clustering of IoT malware behaviors through system call monitoring. Our method is constructed from multiple one-class SVM classifiers and has the ability to classify known malware with F1-Score over 98% and probability to detect unknown malware up to 97%. Unknown malware instances with similar behaviors can also be grouped together so new classes of malware will be discovered. %0 Conference Paper %A Hoang, Dang Kien %A Nguyen, Dai Tho %A Vu, Duy Loi %A Trường Đại học Công nghệ, Đại học Quốc gia Hà Nội, %B Hội thảo Quốc gia lần thứ XXII Một số vấn đề chọn lọc của Công nghệ thông tin và Truyền thông %C Thái Bình %D 2019 %F SisLab:3548 %P 211-215 %T Phân loại hành vi phần mềm độc hại trên các thiết bị IoT dựa vào system call %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3548/ %X Phần mềm độc hại từ lâu đã là một trong các mối nguy hại lớn nhất đối với các hệ thống máy tính, nay đã chuyển sang cả các thiết bị IoT. Trong bài báo này chúng tôi đề xuất sử dụng thuật toán FRNN để thay thế cho thuật toán 1-NN trong quy trình đề xuất bởi Rieck và các cộng sự [2]. Kết quả thực nghiệm cho thấy phương pháp của chúng tôi đề xuất mang lại hiệu quả phân lớp cao đối với các mã độc đã biết với độ đo micro-F_1 lên tới trên 98% trong khi vẫn duy trì khả năng phát hiện các loại mã độc mới lên tới 97%, khả năng chống lại ảnh hưởng của dữ liệu nhiễu so với phương pháp gốc được tăng lên đáng kể. %0 Conference Paper %A Hoang, Duc Tam %A Nguyen, Le Minh %A Pham, Bao Son %B KSE: the 2015 International Conference on Knowledge and Systems Engineering %C Ho Chi Minh city, Vietnam %D 2015 %F SisLab:1606 %T L2S: Transforming natural language questions into SQL queries %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1606/ %X The reliability of a question answering system is bounded by the availability of resources and linguistic tools. In this paper, we introduce a hybrid approach to transforming natural language questions into structured queries. It alleviates the lack of experts in domain observation and the deficient performance of linguistic tools. Specifically, we exploit the semantic information for mapping natural language terminologies to structured query and bipartite graph model for the matching phase. Experimental results on the Vietnam national university entrance exam dataset and the Geoqueries880 dataset achieve accuracies of 91.14% and 87.55%, respectively. %0 Conference Paper %A Hoang, Minh Duong %A Le, Khanh Trinh %A Pham, Ngoc Hung %B 2013: IEEE RIVF International Conference on Computing and Communication Technologies (RIVF2013) %C Hanoi, VIETNAM %D 2013 %F SisLab:768 %P 22-26 %T An Assume-Guarantee Model Checker for Component-Based Systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/768/ %0 Journal Article %@ 1756-3305 %A Hoang, Phuc %A Teo, Tze Min %A Ho, Thien Xuan %A Vinh, Le Sy %D 2016 %F SisLab:2069 %I BioMed %J Parasites & Vectors %N 49 %T Mechanisms of sex determination and transmission ratio distortion in Aedes aegypti %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2069/ %V 9 %X Background: More effective mosquito control strategies are urgently required due to the increasing prevalence of insecticide resistance. The sterile insect technique (SIT) and the release of insects carrying a dominant lethal allele (RIDL) are two proposed methods for environmentally-friendly, species-targeted population control. These methods may be more suitable for developing countries if producers reduce the cost of rearing insects. The cost of control programs could be reduced by producing all-male mosquito populations to circumvent the isolation of females before release without reducing male mating competitiveness caused by transgenes. Results: An RNAi construct targeting the RNA recognition motif of the Aedes aegypti transformer-2 (tra-2) gene does not trigger female-to-male sex conversion as commonly observed among dipterous insects. Instead, homozygous insects show greater mortality among m-chromosome-bearing sperm and mm zygotes, yielding up to 100 % males in the subsequent generations. The performance of transgenic males was not significantly different to wild-type males in narrow-cage competitive mating experiments. Conclusion: Our data provide preliminary evidence that the knockdown of Ae. aegypti tra-2 gene expression causes segregation distortion acting at the level of gametic function, which is reinforced by sex-specific zygotic lethality. This finding could promote the development of new synthetic sex distorter systems for the production of genetic sexing mosquito strains. Keywords: Aedes aegypti, Sex determination, Transformer-2, Meiotic drive, Spermatogenesis, Culicinae %0 Conference Paper %A Hoang, Thi Diep %A Arndt, von Haesele %A Bui, Quang Minh %B 1st Greifswald Phylogenetics Meeting %C Germany %D 2014 %F SisLab:1189 %T Ultrafast Parsimony Bootstrap %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1189/ %0 Journal Article %@ 0737-4038 %A Hoang, Thi Diep %A Chernomor, Olga %A Arndt, von Haeseler %A Bui, Quang Minh %A Le, Sy Vinh %D 2018 %F SisLab:3437 %I Oxford University Press %J Molecular Biology and Evolution %N 2 %P 518-522 %T UFBoot2: Improving the Ultrafast Bootstrap Approximation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3437/ %V 35 %X Abstract The standard bootstrap (SBS), despite being computationally intensive, is widely used in maximum likelihood phylogenetic analyses. We recently proposed the ultrafast bootstrap approximation (UFBoot) to reduce computing time while achieving more unbiased branch supports than SBS under mild model violations. UFBoot has been steadily adopted as an efficient alternative to SBS and other bootstrap approaches. Here, we present UFBoot2, which substantially accelerates UFBoot and reduces the risk of overestimating branch supports due to polytomies or severe model violations. Additionally, UFBoot2 provides suitable bootstrap resampling strategies for phylogenomic data. UFBoot2 is 778 times (median) faster than SBS and 8.4 times (median) faster than RAxML rapid bootstrap on tested data sets. UFBoot2 is implemented in the IQ-TREE software package version 1.6 and freely available at http://www.iqtree.org. %0 Conference Paper %A Hoang, Thi Diep %A Kaur, Jasleen %A Menczer, Filippo %B WebSci10: Extending the Frontiers of Society On-Line %C 26-27 April 2010 %D 2010 %F SisLab:2356 %T Crowdsourcing Scholarly Data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2356/ %0 Journal Article %@ 1471-2148 %A Hoang, Thi Diep %A Le, Sy Vinh %A Flouri, Tomas %A Stamatakis, Alexandros %A von Haeseler, Arndt %A Bui, Quang Minh %D 2018 %F SisLab:2955 %J BMC Evolutionary Biology %T MPBoot: fast phylogenetic maximum parsimony tree inference and bootstrap approximation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2955/ %V 18 %X BACKGROUND: The nonparametric bootstrap is widely used to measure the branch support of phylogenetic trees. However, bootstrapping is computationally expensive and remains a bottleneck in phylogenetic analyses. Recently, an ultrafast bootstrap approximation (UFBoot) approach was proposed for maximum likelihood analyses. However, such an approach is still missing for maximum parsimony. RESULTS: To close this gap we present MPBoot, an adaptation and extension of UFBoot to compute branch supports under the maximum parsimony principle. MPBoot works for both uniform and non-uniform cost matrices. Our analyses on biological DNA and protein showed that under uniform cost matrices, MPBoot runs on average 4.7 (DNA) to 7 times (protein data) (range: 1.2-20.7) faster than the standard parsimony bootstrap implemented in PAUP*; but 1.6 (DNA) to 4.1 times (protein data) slower than the standard bootstrap with a fast search routine in TNT (fast-TNT). However, for non-uniform cost matrices MPBoot is 5 (DNA) to 13 times (protein data) (range:0.3-63.9) faster than fast-TNT. We note that MPBoot achieves better scores more frequently than PAUP* and fast-TNT. However, this effect is less pronounced if an intensive but slower search in TNT is invoked. Moreover, experiments on large-scale simulated data show that while both PAUP* and TNT bootstrap estimates are too conservative, MPBoot bootstrap estimates appear more unbiased. CONCLUSIONS: MPBoot provides an efficient alternative to the standard maximum parsimony bootstrap procedure. It shows favorable performance in terms of run time, the capability of finding a maximum parsimony tree, and high bootstrap accuracy on simulated as well as empirical data sets. MPBoot is easy-to-use, open-source and available at http://www.cibiv.at/software/mpboot . %0 Journal Article %A Hoang, Thi Diep %A Le, Sy Vinh %A Tomas, Flouri %A Alexandros, Stamatakis %A Arndt, von Haeseler %A Minh, Bui %D 2018 %F SisLab:3310 %J BMC Evolutionary Biology, 18:11 %N 18 %P 11 %T Fast phylogenetic maximum parsimony tree inference and bootstrap approximation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3310/ %X Background: The nonparametric bootstrap is widely used to measure the branch support of phylogenetic trees. However, bootstrapping is computationally expensive and remains a bottleneck in phylogenetic analyses. Recently, an ultrafast bootstrap approximation (UFBoot) approach was proposed for maximum likelihood analyses. However, such an approach is still missing for maximum parsimony. Results: To close this gap we present MPBoot, an adaptation and extension of UFBoot to compute branch supports under the maximum parsimony principle. MPBoot works for both uniform and non-uniform cost matrices. Our analyses on biological DNA and protein showed that under uniform cost matrices, MPBoot runs on average 4.7 (DNA) to 7 times (protein data) (range: 1.2–20.7) faster than the standard parsimony bootstrap implemented in PAUP*; but 1.6 (DNA) to 4.1 times (protein data) slower than the standard bootstrap with a fast search routine in TNT (fast-TNT). However, for non-uniform cost matrices MPBoot is 5 (DNA) to 13 times (protein data) (range:0.3–63. 9) faster than fast-TNT. We note that MPBoot achieves better scores more frequently than PAUP* and fast-TNT. However, this effect is less pronounced if an intensive but slower search in TNT is invoked. Moreover, experiments on large-scale simulated data show that while both PAUP* and TNT bootstrap estimates are too conservative, MPBoot bootstrap estimates appear more unbiased. Conclusions: MPBoot provides an efficient alternative to the standard maximum parsimony bootstrap procedure. It shows favorable performance in terms of run time, the capability of finding a maximum parsimony tree, and high bootstrap accuracy on simulated as well as empirical data sets. MPBoot is easy-to-use, open-source and available at http://www.cibiv.at/software/mpboot. %0 Conference Paper %A Hoang, Thi Diep %A Menczer, Filippo %B SUNBELT XXX %C Riva del Garda Fiere Congressi Riva del Garda (TN), Italy %D 2010 %F SisLab:2357 %T Building interdisciplinary collaboration networks from a social citation analysis tool %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2357/ %0 Conference Paper %A Hoang, Thi Diep %A Pham, Hong Nguyen %A Dinh, Ngoc Thang %A Le, Hai Yen %B The 8th National Conference on Communication and Information Technology %D 2005 %F SisLab:2154 %T Evaluating Approaches to Word Segmentation for Vietnamese Text %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2154/ %0 Conference Paper %A Hoang, Thi Diep %A Tran, Ke %A Nguyen, Le Minh %A Ke Tran, %A Le-Minh Nguyen, %B ITC.rda’08 National Symposium on Research, Development and Application of Information and Communication %D 2008 %F SisLab:2155 %T Revision Rules Combine with MST Method for Vietnamese Dependency Parsing %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2155/ %0 Conference Paper %A Hoang, Thi Diep %A Vinh, Le Sy %A Flouri, Tomas %A Stamatakis, Alexandros %A von Haeseler, Arndt %A Minh, Bui Quang %B 2016 Eighth International Conference on Knowledge and Systems Engineering (KSE) %D 2016 %F SisLab:2355 %P 1-6 %T A new phylogenetic tree sampling method for maximum parsimony bootstrapping and proof-of-concept implementation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2355/ %X Reconstructing maximum parsimony (MP) trees along with bootstrap for large multiple sequence alignments is a computationally challenging task. In this paper, we present MPBoot, a new method for approximating MP bootstrap based on the idea of a previously introduced method for maximum likelihood bootstrap. MPBoot economizes the computations by only sampling trees on the original alignment and simultaneously calculating exact parsimony scores for all bootstrap alignments. Preliminary experiments on simulated data showed that MPBoot is about 2-3.5 times faster and yields better parsimony scores than a fast bootstrap procedure implemented in TNT. MPBoot also produced less biased support value than TNT for small simulated data sets (<;= 200 sequences). However, for large number of sequences (>=500) MPBoot overestimates the probability that a clade is true, i.e., results in likely high false positive rates. As MPBoot is currently a proof-of-concept implementation, more work is to understand the limitations of the method. %0 Conference Paper %A Hoang, Thi Diep %A von Haeseler, Arndt %A Bui, Quang Minh %B Greifswald Phylogenetics Meeting 2014 %C Greifswald Germany %D 2014 %F SisLab:2358 %T Ultrafast Parsimony Bootstrap %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2358/ %0 Conference Paper %A Hoang, Thi Linh %A Pham, Tuan Dung %A Ta, Viet Cuong %B The 13th International Conference on KNOWLEDGE AND SYSTEMS ENGINEERING (KSE 2021) %C Bangkok, Thailand %D 2021 %F SisLab:4688 %T Improving Graph Convolutional Networks with Transformer Layer in social-based items recommendation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4688/ %X With the emergence of online social networks, social-based items recommendation has become a popular research direction. Recently, Graph Convolutional Networks have shown promising results by modeling the information diffusion process in graphs. It provides a unified framework for graph embedding that can leverage both the social graph structure and node features information. In this paper, we improve the embedding output of the graph-based convolution layer by adding a number of transformer layers. The transformer layers with attention architecture help discover frequent patterns in the embedding space which increase the predictive power of the model in the downstream tasks. Our approach is tested on two social-based items recommendation dataset, Ciao and Epinions and our model outperforms other graph-based recommendation baselines. %0 Conference Paper %A Hoang, Thi Linh %A Ta, Viet Cuong %B 19th Pacific Rim International Conference on Artificial Intelligence %F SisLab:4776 %T Dynamic-GTN: Learning an Node Efficient Embedding in Dynamic Graph with Transformer %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4776/ %X Graph Transformer Networks (GTN) use attention mechanism to learn the node representation in a static graph and can achieve state-of-the-art results on several graph learning tasks. However, due to the computation complexity of the attention operation, these models are not applicable to dynamic graphs. In this paper, we propose the Dynamic-GTN model which is designed to learn the node embedding in a continous-time dynamic graph. The Dynamic-GTN extends the attention mechanism in a standard GTN to include temporal information of recent node interaction. Based on temporal patterns interaction between nodes, an node sampling step is added to reduce the number of attention connections in the dynamic graph. We evaluate our model on three benchmark datasets for learning node embedding in dynamic graphs. The results show that the Dynamic-GTN has better accuracy than the state-of-the-art of graph neural networks on both transductive and inductive graph learning tasks. %0 Conference Paper %A Hoang, Tieu Binh %A Ma, Thi Chau %A Akihiro, Sugimoto %A Bui, The Duy %B 2018 the 7th International Conference on Computer and Communication Engineering (ICCCE) %C Malaysia %D 2018 %F SisLab:3123 %T Selecting active frames for action recognition with 3D convolutional network %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3123/ %X Recent applications of Convolutional Neural Networks, especially 3-Dimensional Convoltutional Neural Networks (3DCNNs) for human action recognition (HAR) in videos have widely used. In this paper, we use a multi-stream framework which is a combination from separated networks with different kind of input generated from unique video dataset. To achieve the high results, firstly, we proposed a method to extract the active frames (called Selected Active Frames - SAF) from a videos to build datasets for 3DCNNs in video classifying problem. Second, we deploy a new approach called Vote fusion which considered as an effective fusion method for ensembling multi-stream networks. From the various datasets generated from videos, we extract frames by our method and feed into 3DCNNs for feature extraction, then we carry out training and then fuse the results of softmax layers of these streams. We evaluate the proposed methods on solving action recognition problem. These method are carried on three well-known datasets (HMFB51, UCF101, and KTH). The results are also compared to the state-of-the-art results to illustrate the efficiency and effectiveness in our approach %0 Conference Paper %A Hoang, Trung Kien %A Nguyen, Kim Hung %A Ma, Thi Chau %A Ngo, Thi Duyen %A Nguyen, Xuan Thanh %B The 2017 International Conference on Knowledge and Systems Engineering (KSE) %C Hue, Vietnam %D 2017 %F SisLab:2648 %T Single View Image Based – 3D Human Pose Reconstruction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2648/ %0 Conference Paper %A Hoang, Trung Kien %A Nguyen, Kim Hung %A Nguyen, Xuan Thanh %A Ma, Thi Chau %A Ngo, Thi Duyen %B The 2017 International Conference on Knowledge and Systems Engineering (KSE) %C Hue, Vietnam %D 2017 %F SisLab:2478 %T Single View Image Based - 3D Human Pose Reconstruction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2478/ %0 Conference Paper %A Hoang, Van Truong %A Phung, Manh Duong %B The International Conference on Advanced Mechanical Engineering, Automation, and Sustainable Development 2021 (AMAS2021) %D 2022 %F SisLab:4741 %P 743-753 %T Enhanced Teaching-Learning-based Optimization for 3D Path Planning of Multicopter UAVs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4741/ %X This paper introduces a new path planning algorithm for unmanned aerial vehicles (UAVs) based on the teaching-learning-based optimization (TLBO) technique. We first define an objective function that incorporates requirements on the path length and constraints on the movement and safe operation of UAVs to convert the path planning into an optimization problem. The optimization algorithm named Multi-subject TLBO is then proposed to minimize the formulated objective function. The algorithm is developed based on TLBO but enhanced with new operations including mutation, elite selection and multi-subject training to improve the solution quality and speed up the convergence rate. Comparison with state-of-the-art algorithms and experiments with real UAVs have been conducted to evaluate the performance of the proposed algorithm. The results confirm its validity and effectiveness in generating optimal, collision-free and flyable paths for UAVs in complex operating environments. %0 Conference Paper %A Hoang, Van Xiem %A Dinh, Trieu Duong %A Le, Thanh Ha %B the 2016 International Conference Advanced Technologies for Communications (ATC) %C Hanoi, Vietnam %D 2016 %F SisLab:1992 %P 1-6 %T Spatial-Temporal Feature Extraction based Adaptive Search Range for Effective Frame Rate-Up Conversion %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1992/ %X Frame rate up conversion (FRUC) has been playing an important role in the recent development of advanced television systems due to its benefits on both increasing the viewing quality experiences and reducing the cost of video transmission. However, with the increasing of video resolutions, notably from Standard Definition (SD) to High Definition (HD), FRUC is now asked to provide not only better interpolated frames quality but also lower time processing. Therefore, in this paper, we propose a novel spatial – temporal feature extraction based adaptive search range for effective FRUC. In the proposed adaptive search range scheme, a set of temporal and spatial features are carefully defined and exploited to adaptively assign an appropriate search range value to each considered block; thus, directly reducing the FRUC time processing. Moreover, since the optimal search range can be employed; the quality of interpolated frames is significantly improved. Experimental results conducted for a rich set of video test sequences shown the advantages of the proposed FRUC scheme, notably in both subjective objective image quality improvement and time processing reduction. %0 Journal Article %@ 0866-8612 %A Hoang, Van Xiem %A duong, thi Hang %A Trinh, Anh Vu %A Vu, Xuan Thang %D 2019 %F SisLab:3456 %I VNU Journal of Science %J VNU Journal of Computer Science and Communication Engineering %N 1 %P 1-9 %T Cooperative Caching in Two-Layer Hierarchical Cache aided Systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3456/ %V 35 %X Caching has received much attention as a promising technique to overcome high data rate and stringent latency requirements in the future wireless networks. The premise of caching technique is to prefetch most popular contents closer to end users in local cache of edge nodes, e.g., base station (BS). When a user requests a content that is available in the cache, it can be served directly without being sent from the core network. In this paper, we investigate the performance of hierarchical caching systems, in which both BS and end users are equipped with a storage memory. In particular, we propose a novel cooperative caching scheme that jointly optimizes the content placement at the BS’s and users’ caches. The proposed caching scheme is analytically shown to achieve a larger global caching gain than the reference in both uncoded – and coded caching strategies. Finally, numerical results are presented to demonstrate the effectiveness of our proposed caching algorithm. %0 Journal Article %@ 1999-4893 %A Hoang, Xuan Huan %A Pham, Van Canh %A Ha, K.T. Dung %A Vu, C Quang %A Nguyen, Su Anh %A Canh V. Pham, %A Dung K. T. Ha, %A Quang C. Vu, %A Anh N. Su, %D 2020 %F SisLab:4073 %I MDPI %J MDPI algorithms %N 8 (183 %P 1-23 %T Influence Maximization with Priority in Online Social Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4073/ %V 13 %X The Influence Maximization (IM) problem, which finds a set of k nodes (called seedset) in a social network to initiate the influence spread so that the number of influenced nodes after propagation process is maximized, is an important problem in information propagation and social network analysis. However, previous studies ignored the constraint of priority that led to inefficient seed collections. In some real situations, companies or organizations often prioritize influencing potential users during their influence diffusion campaigns. With a new approach to these existing works, we propose a new problem called Influence Maximization with Priority (IMP) which finds out a set seed of k nodes in a social network to be able to influence the largest number of nodes subject to the influence spread to a specific set of nodes U (called priority set) at least a given threshold T in this paper. We show that the problem is NP-hard under well-known IC model. To find the solution, we propose two efficient algorithms, called Integrated Greedy (IG) and Integrated Greedy Sampling (IGS) with provable theoretical guarantees. IG provides a � 1 − (1 − 1 k ) t � -approximation solution with t is an outcome of algorithm and t ≥ 1. The worst-case approximation ratio is obtained when t = 1 and it is equal to 1/k. In addition, IGS is an efficient randomized approximation algorithm based on sampling method that provides a � 1 − (1 − 1 k ) t − e � -approximation solution with probability at least 1 − δ with e > 0, δ ∈ (0, 1) as input parameters of the problem. We conduct extensive experiments on various real networks to compare our IGS algorithm to the state-of-the-art algorithms in IM problem. The results indicate that our algorithm provides better solutions interns of influence on the priority sets when approximately give twice to ten times higher than threshold T while running time, memory usage and the influence spread also give considerable results compared to the others. %0 Report %9 Technical Report %@ UET-FIT-2018xyz %A Hoang, Xuan Tung %D 2018 %F SisLab:3343 %I VNU University of Engineering and Technology %T Function-based Semantic-ware Cache Replacement Algorithm for Web Systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3343/ %X As the web expands overwhelmingly in our daily lives, the pressure to improve the performance of web servers increases. A significant optimization technique that enables scalable web servers to serve clients more efficiently and with lower resource demands consists in caching requested web objects on intermediate cache servers. At the core of the cache server operation is the replacement algorithm, which is responsible for selecting, according to a cache replacement policy, the cached pages that should be removed in order to make space for new pages. Traditional replacement policies used in practice take advantage of temporal reference locality by removing the least recently/frequently requested pages from the cache. Lately, due to the properties of semantic content in of web page, there are some new attempts to implement cache replacement based on multiple parameters including semantic content. This report presents a semantic-aware caching policy enhanced with the use of function-based cache value. Our algorithm is believed to outperform traditional methods in terms of hit rate, which can be useful for website with many small and equal-size web objects. %0 Conference Paper %A Hoang, Xuan Tung %A Bui, Ngoc Dung %B RIVF: the 2019 IEEE-RIVF International Conference on Computing and Communication Technologies %C Da Nang, Vietnam %D 2019 %F SisLab:3837 %P 142-147 %T An Enhanced Semantic-based Cache Replacement Algorithm for Web Systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3837/ %X As Web traffics is increasing on the Internet, caching solutions for Web systems are becoming more important since they can greatly expand system scalability. An important part of a caching solution is cache replacement policy, which is responsible for selecting victim items that should be removed in order to make space for new objects. Typical replacement policies used in practice only take advantage of temporal reference locality by removing the least recently/frequently requested items from the cache. Although those policies work well in memory or filesystem cache, they are inefficient for Web systems since they do not exploit semantic relationship between Web items. This paper presents a semantic-aware caching policy that can be used in Web systems to enhance scalability. The proposed caching mechanism defines semantic distance from a web page to a set of pivot pages and use the semantic distances as a metric for choosing victims. Also, it use a function-based metric that combines access frequency and cache item size for tie-breaking. Our simulations show that out enhancements outperform traditional methods in terms of hit rate, which can be useful for websites with many small and similar-in-size web objects. %0 Conference Paper %A Hoang, Xuan-Huan %A Duong, T.A. Tuyet %A Doan, T.T. Ha %A Nguyen, T. Hung %B 2014: The 6th International Conference on Knowledge and Systems Engineering (KSE 2014) %C Hanoi, Vietnam %D 2014 %F SisLab:446 %P 589-601 %T An Efficient Ant Colony Algorithm for DNA Motif Finding %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/446/ %0 Conference Paper %A Hoang, Xuan-Tung %B HỘI THẢO KHOA HỌC MÁY TÍNH VÀ CÁC HỆ THỐNG THÔNG MINH %C Hanoi %D 2018 %F SisLab:3344 %T Cơ chế đồng bộ hoá liên lược đồ cho các ứng dụng cơ sở dữ liệu phân tán %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3344/ %X Đồng bộ hoá cơ sở dữ liệu là một trong những tính năng quan trọng trong các hệ thống tính toán phân tán. Đồng bộ hoá liên lược đồ (cross-schema) cho phép việc đồng bộ hoá có thể thực hiện được trong môi trường cơ sở dữ liệu không đồng nhất nơi mà các cơ sở dữ liệu có cấu trúc khác nhau vẫn thực hiện đồng bộ dữ liệu được với nhau. Trong bài báo này, chúng tôi giới thiệu XSync, một middleware được thiết kế nhằm phục vụ thực hiện việc đồng bộ hoá liên lược đồ một cách hiệu quả cho các hệ thống sử dụng cơ sở dữ liệu. Với XSync, nhà phát triển phần mềm có thể dễ dàng tích hợp khả năng đồng bộ dữ liệu vào hệ thống IT của mình một cách linh hoạt với chi phí phát triển thấp nhưng vẫn đem lại nhiều lợi ích lớn về mặt tính năng cũng như hiệu năng. %0 Journal Article %A Hoang Nhu, Dong %A Nguyen Nam, Hoang %A Hoang Trong, Minh %A Saba, Takahiko %D 2017 %F SisLab:2631 %J the Research and Development on Information and Communication Technology Journal %N 14 %P 45-51 %T Femtocell Selection Scheme for Reducing Unnecessary Handover and Enhancing Downlink QoS in Cognitive Femtocell Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2631/ %V E3 %X Femtocell networks have been proposed for indoor communications as the extension of cellular networks for enhancing coverage performance. Because femtocells have small coverage radius, typically from 15 to 30 meters, a femtocell user (FU) walking at low speed can still make several femtocell-to-femtocell handovers during its connection. When performing a femtocell-to-femtocell handover, femtocell selection used to select the target handover femtocell has to be able not only to reduce unnecessary handovers and but also to support FU’s quality of service (QoS). In the paper, we propose a femtocell selection scheme for femtocell-tofemtocell handover, named Mobility Prediction and Capacity Estimation based scheme (MPCE-based scheme), which has the advantages of the mobility prediction and femtocell’s available capacity estimation methods. Performance results obtained by computer simulation show that the proposed MPCE-based scheme can reduce unnecessary femtocell-tofemtocell handovers, maintain low data delay and improve the throughput of femtocell users. %0 Conference Paper %A Hussein, Shamsul Faisal Mohd %A Bakar, Mohd Anuar Abu %A Makino, Yoshiki %A Nguyen, Hoai Son %A Abdullah, Shahrum Shah %A Yuto, Lim %A Yasuo, Tan %B 17th Asia Simulation Conference (AsiaSim) %C Malaysia %D 2017 %F SisLab:2659 %T Simplifying the Auto Regressive and Moving Average (ARMA) Model - Representing the Dynamic Thermal Behaviour of iHouse Based on Theoretical Knowledge %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2659/ %X Modelling and simulation is an alternative way of testing the dynamic behaviour of a real system – in some situation, testing the real system are expensive, time consuming, not comfortable, and dangerous. Mathematical model describing the dynamic behaviour of a system can be represented by using white, black, or grey box model. This study focuses on developing a simplified Auto Regressive Moving Average (ARMA) model (a type of linear black model) to represent the dynamic thermal behaviour of iHouse – simpli- fication is done based on the theoretical knowledge of the building. The per- formance of the simplified ARMA model developed in this study is compared with the performance of the models developed in previous studies, which are: (1) House Thermal Simulator; (2) and ARMA model. Result shows that the simplified ARMA model developed in this study consists of simpler set of mathematical equations, but can still simulate the dynamic thermal behaviour of iHouse with the accuracy that is almost on par with the models developed in previous studies. %0 Journal Article %@ 0098-3063 %A Hwang, Min-Cheol %A Le, Thanh Ha %A Kim, Nam-hyeong %A Park, Chun-Su %A Ko, Sung-Jea %D 2007 %F SisLab:1906 %I IEEE %J IEEE Transactions on Consumer Electronics %N 1 %P 218-226 %T Person identification system for future digital TV with intelligence %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1906/ %V 53 %0 Conference Paper %A Ikechukwu, Ogbodo Mark %A Dang, Nam Khanh %A Abdallah, Abderazek Ben %B International Symposium on Ubiquitous Networking 2021 %D 2021 %F SisLab:4618 %T Energy-efficient Spike-based Scalable Architecture for Next-generation Cognitive AI Computing Systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4618/ %0 Conference Paper %A Ivanov\', Marika %A Surynek, Pavel %A Nguyen, Diep Thi Ngoc %B the Thirty-First International Flairs Conference %D 2017 %F SisLab:3793 %P 348-353 %T Maintaining Ad-Hoc Communication Network in Area Protection Scenarios with Adversarial Agents %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3793/ %V abs/17 %X We address a problem of area protection in graph-based scenarios with multiple mobile agents where connectivity is maintained among agents to ensure they can communicate. The problem consists of two adversarial teams of agents that move in an undirected graph shared by both teams. Agents are placed in vertices of the graph; at most one agent can occupy a vertex; and they can move into adjacent vertices in a conflict free way. Teams have asymmetric goals: the aim of one team - attackers - is to invade into given area while the aim of the opponent team - defenders - is to protect the area from being entered by attackers by occupying selected vertices. The team of defenders need to maintain connectivity of vertices occupied by its own agents in a visibility graph. The visibility graph models possibility of communication between pairs of vertices. We study strategies for allocating vertices to be occupied by the team of defenders to block attacking agents where connectivity is maintained at the same time. To do this we reserve a subset of defending agents that do not try to block the attackers but instead are placed to support connectivity of the team. The performance of strategies is tested in multiple benchmarks. The success of a strategy is heavily dependent on the type of the instance, and so one of the contributions of this work is that we identify suitable strategies for diverse instance types. %0 Journal Article %@ 1932-6203 %A Kaur, Jasleen %A Hoang, Thi Diep %A Xiaoling Sun, %A Lino Possamai, %A Mohsen Jafari Asbagh, %A Snehal Patil, %A Filippo Menczer, %D 2012 %F SisLab:2159 %J PLOS ONE %T Scholarometer: A Social Framework for Analyzing Impact across Disciplines %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2159/ %0 Conference Paper %A Khuat, Thanh Son %A Nguyen, Trường Thang %A Tran, Hung %A Nguyen, Ha Thanh %B Hội thảo Quốc gia thường niên "Một số vấn đề chọn lọc của Công nghệ thông tin và Truyền thông" %C Thanh Hoa, Vietnam %D 2018 %F SisLab:3125 %T Công nghệ Blockchain trong vấn đề đảm bảo tính toàn vẹn và minh bạch của dữ liệu %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3125/ %X Báo cáo này tiếp cận tìm hiểu công nghệ Blockchain – một công nghệ đang phát triển mạnh mẽ và bước đầu ứng dụng trong nhiều lĩnh vực. Dựa trên bản chất bảo mật dữ liệu của Blockchain, chúng tôi đề xuất ứng dụng cho bài toán bảo vệ tính toàn vẹn và minh bạch cho dữ liệu và cụ thể ở đây là các dữ liệu văn bản. Trước hết, báo cáo trình bày khái quát về công nghệ thành phần trong Blockchain và sự phù hợp của nó đối với bài toán xác thực/bảo vệ dữ liệu. Tiếp theo, báo cáo đê xuất giải pháp ứng dụng Blockchain trong bài toán thực tế liên quan tới tính toàn vẹn và tính minh bạch của tài liệu, đánh giá mức độ an toàn của việc áp dụng. %0 Conference Paper %A Kieu, Minh Viet %A Nguyen, Dai Tho %A Nguyen, Thanh Thuy %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1523 %T Chống tấn công từ chối dịch vụ tốc độ thấp vào giao thức TCP bằng các cải tiến cơ chế quản lý hàng đợi tích cực. %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1523/ %X Tấn công từ chối dịch vụ tốc độ thấp vào giao thức TCP (LDoS) là hình thức tấn công từ chối dịch vụ phân tán (DDoS) mới hiện nay, đã được giới thiệu lần đầu tiên bởi Aleksandar Kuzmanovic và Edward W. Knightly năm 2003. LDoS khai thác điểm yếu của giao thức TCP đó là sử dụng giá trị thời gian chờ phát lại gói tin nhỏ nhất minRTO là hằng số bằng 1 giây. Chống LDoS trở nên khó khăn hơn so với chống DDoS thông thường vì LDoS gửi gói tin với tốc độ trung bình thấp nên nó tránh được sự phát hiện của các bộ giám sát mạng (chủ yếu phát hiện bất thường dựa trên sự bùng nổ gói tin vào mạng). Nghiên cứu đưa ra độ đo mới Congestion Participation Rate (CPR) để phân biệt dòng TCP thông thường và dòng tấn công LDoS, cải tiến thuật toán quản lý hàng đợi RED (Random Early Detection) để chống LDoS. Kết quả mô phỏng bằng phần mềm NS-2 cho thấy chỉ số CPR của dòng TCP thường nhỏ hơn so với chỉ số CPR của dòng tấn công LDoS; trên cơ sở đó, thuật toán loại bỏ gói tin của các dòng có chỉ số CPR cao và cho đi qua gói tin của các dòng có chỉ số CPR thấp. %0 Conference Paper %A Kieu, Minh Viet %A Nguyen, Dai Tho %A Nguyen, Thanh Thuy %A University of Engineering and Technology, Vietnam National University, Hanoi, %B SoICT 2018: 9th International Symposium on Information and Communications %C Da Nang %D 2018 %F SisLab:3378 %P 163-168 %T Techniques for Improving Performance of the CPR-Based Approach %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3378/ %X TCP-targeted low-rate distributed denial-of-service (LDDoS) attacks have created an opportunity for attackers to reduce their total attaking rate (and hence, the detection probability of the attacks) while inflicting the same damage to TCP flows as traditional flooding-based DDoS attacks. CPR-based approach has been proposed by Zhang et al. to detect and filter this kind of DDoS attacks,but its performance in terms of TCP throughput under attack is shown to be limited by the way it calculates CPR for each flow. In this paper, we will propose some modifications to the CPR-based approach in order to increase its performance. Simulation results show that the modifications can increase performance significantly %0 Conference Paper %A Kieu, Minh Viet %A Nguyen, Dai Tho %A Nguyen, Thanh Thuy %A University of Engineering and Technology, Vietnam National University, Hanoi, %B The Eighth International Symposium on Information and Communication Technology (SoICT 2017) %C Nha Trang, Vietnam %D 2017 %F SisLab:2752 %T Using CPR Metric to Detect and Filter Low-Rate DDoS Flows %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2752/ %X TCP-targeted low-rate distributed denial-of-service (LDDoS) attacks pose a serious challenge to the reliability and security of the Internet. Among various proposed solutions, we are particularly interested in the Congestion Participation Rate (CPR) metric and the CPR-based approach. Through a simulation study, we show that the existing algorithm cannot simultaneously achieve high TCP throughput while under attack and good fairness performance for new legitimate TCP flows in normal times. We then propose a new version of the CPR-based approach to overcome the tradeoff. Simulation results show that it preserves TCP throughput while under attack fairly well, yet maintains fairness for new TCP flows in normal times. %0 Conference Paper %A Kieu, Minh Viet %A Nguyen, Dai Tho %A Nguyen, Thanh Thuy %A VNU University of Engineering and Technology, %B The 2020 RIVF International Conference on Computing and Communication Technologies (RIVF 2020) %C Ho Chi Minh City, Vietnam %D 2002 %F SisLab:3999 %T A Way to Estimate TCP Throughput under Low-Rate DDoS Attacks: One TCP Flow %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3999/ %X TCP-targeted low-rate distributed denial-of-service (LDDoS) attacks were first introduced by A. Kuzmanovic and E. Knightly in 2003. The authors also proposed a simple model to quantify TCP throughput under LDDoS attacks. Since then, there have been many researchers attemping to estimate the throughput, such as Luo et al. We agree with them upon the sketch of TCP congestion window under a successful LDDoS attack but we find out that there are more cases than what has been specified. Moreover, the relative error of Luo’s estimation method is still high. Our goal in this paper is to propose a simple but more accurate method to estimate TCP throughput of a single TCP flow under such DDoS attacks. Our estimation values in various scenarios are compared with the results of simulations performed with NS-2 simulator, so that the effectiveness of our method is illustrated. %0 Conference Paper %A Kieu, Thanh Binh %A Pham, Bao Son %A Phan, Xuan Hieu %A Piccardi, Massimo %B The 16th International Conference of the Pacific Association for Computational Linguistics (PACLING 2019) %C Hanoi, Vietnam %D 2019 %F SisLab:3639 %T A Submodular Approach for Reference Recommendation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3639/ %X Choosing appropriate references for a given topic is an important, yet challenging task. The pool of potential candidates is typically very large, in the order of tens of thousands, and growing by the day. For this reason, this paper proposes an approach for automatically providing a reference list for a given abstract. The approach is based on an original submodular inference function which balances relevance, coverage and diversity in the reference list. Experiments are carried out using an ACL corpus as a source for the references and evaluated in terms of precision-recall, MAP and MRR. The results show the remarkable comparative performance of the proposed approach. %0 Conference Paper %A Kieu, Thanh Binh %A Unanue, Inigo Jauregi %A Pham, Bao Son %A Phan, Xuan Hieu %A Piccardi, Massimo %B The 25th International Conference on Pattern Recognition %C Milan, Italy %D 2020 %F SisLab:4216 %T Learning Neural Textual Representations for Citation Recommendation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4216/ %X With the rapid growth of the scientific literature, manually selecting appropriate citations for a paper is becoming increasingly challenging and time-consuming. While several approaches for automated citation recommendation have been proposed in the recent years, effective document representations for citation recommendation are still elusive to a large extent. For this reason, in this paper we propose a novel approach to citation recommendation which leverages a deep sequential representation of the documents (Sentence-BERT) cascaded with Siamese and triplet networks in a submodular scoring function. To the best of our knowledge, this is the first approach to combine deep representations and submodular selection for a task of citation recommendation. Experiments have been carried out using a popular benchmark dataset - the ACL Anthology Network corpus - and evaluated against baselines and a state-of-the-art approach using metrics such as the MRR and F1-at-k score. The results show that the proposed approach has been able to outperform all the compared approaches in every measured metric. %0 Conference Paper %A Kieu, Thanh Binh Thanh Binh %A Unanue, Inigo Jauregi %A Pham, Bao Son %A Phan, Xuan Hieu %A Piccardi, Massimo %B The 25th International Conference on Pattern Recognition %C Milan, Italy %D 2020 %F SisLab:4158 %T Learning Neural Textual Representations for Citation Recommendation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4158/ %X With the rapid growth of the scientific literature, manually selecting appropriate citations for a paper is becoming increasingly challenging and time-consuming. While several approaches for automated citation recommendation have been proposed in the recent years, effective document representations for citation recommendation are still elusive to a large extent. For this reason, in this paper we propose a novel approach to citation recommendation which leverages a deep sequential representation of the documents (Sentence-BERT) cascaded with Siamese and triplet networks in a submodular scoring function. To the best of our knowledge, this is the first approach to combine deep representations and submodular selection for a task of citation recommendation. Experiments have been carried out using a popular benchmark dataset - the ACL Anthology Network corpus - and evaluated against baselines and a state-of-the-art approach using metrics such as the MRR and F1-at-k score. The results show that the proposed approach has been able to outperform all the compared approaches in every measured metric. %0 Book Section %@ %A Kiyoki, Yasushi %A Sasaki, Shiori %A Nguyen, Nhung Trang %A Nguyen, Diep Thi Ngoc %B Conceptual Modelling and Its Theoretical Foundations %D 2012 %F SisLab:3823 %I Springer %P 316-328 %T Cross-Cultural Multimedia Computing with Impression-Based Semantic Spaces %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3823/ %V 7260 %X Over the past decade, the rapid progress of multimedia data management technology has realized the large scale of media data transfer and resource-accumulation in the world. The multimedia computing technology has also been creating new information provision environments in the world-wide scope. Innovative integrations of large scale multimedia data management and computing technology will lead to a new information society. In the design of multimedia systems, one of the most important issues is how to search and analyze media data (images, music, movies and documents), according to impressions and contexts. We have proposed and introduced a "Kansei" and semantic associative search method based on our “Mathematical Model of Meaning (MMM) [11], [13], [14]”. The concept of "Kansei" includes several meanings on sensitive recognition, such as "impression", "human senses", "feelings", "sensitivity", "psychological reaction" and "physiological reaction". This model realizes "Kansei" processing and semantic associative search for media data, according to user’s impressions and contexts. This model is applied to compute semantic correlations between keywords, images, music and documents dynamically in a context-dependent way. The main feature of this model is to realize semantic associative search in the 2000 dimensional orthogonal semantic space with semantic projection functions. This space is created for dynamically computing semantic equivalence or similarity between keywords and media data. We have constructed "Cross-Cultural Multimedia Computing Systems” for sharing and analyzing different cultures with semantic associative functions applied to “cultural & multimedia data,” as a new platform of cross-cultural collaborative environments. This environment enables to create a remote, interactive and real-time cultural and academic research exchange among different countries and cultures. %0 Conference Paper %A Konlakorn, Wongpatikaseree %A Nguyen, Hoai Son %A Yasuo, Tan %B SGSH 2015: Smart Grid and Smart Home %D 2015 %F SisLab:1617 %T Thermal simulation for evaluation of thermal comfort in real houses %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1617/ %0 Conference Paper %A Kosugi, Naoko %A Shimizu, Sachiko %A Sasaki, Shiori %A Nguyen, Diep Thi Ngoc %A Kiyoki, Yasushi %B iiWAS '14 Proceedings of the 16th International Conference on Information Integration and Web-based Applications & Services %C Hanoi Vietnam %D 2014 %F SisLab:3818 %P 385-394 %T Music Retrieval and Adjustment Technique to Support and Motivate Ergotherapy and Daily Exercises %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3818/ %X To maintain and uplift the motivation for doing exercise which is tend to be too simple in rehabilitation and ergotherapy, doing exercise with music is one of the good solutions. In this paper, we confirm that doing exercise with music is fun. Next, we design five types of musical features which relate to exercises. These features are evaluated by doing five types of exercises with listening to the tunes. Based on the results obtained from the experiments, we found many new things, for example, tempo has the adjustment limitation based on how well the user knows the tune. In this paper, we report the latest results from the experiments and explain the way to progress of this research. %0 Conference Paper %A Kubo, Masatoshi %A Nguyen, Hai Chau %A Brik, Sergey %A Yamamoto, Tomoyuki %B 30th annual meeting of Material Research Society of Japan, 2020 (MRS-J 2020) %C online %D 2020 %F SisLab:4234 %T Modelling of Lattice Constants of the Garnet Structured Compounds Using Machine Learning %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4234/ %X Garnet structured compounds have been widely utilized as the host materials of phosphors. Due to their complicated structures with numerous atoms in a unit cell of the garnet structure, it is difficult to predict their crystal structures accurately by the first principles calculation, especially when they include magnetic elements. It was proposed to predict the structural parameters for spinel compounds by the empirical model employing ionic radii and electronegativities, which reproduced the experimental structural parameters of spinels successfully. In the present work, we propose a comprehensive and reliable model to explain and/or predict the structural parameters of the garnet structured materials using the machine learning. The lattice parameters of 182 garnet compounds reported in the database are compiled in our dataset to train the model. Using the ionic radii and electronegativities of the constituent elements of considered garnet compounds, we constructed the linear regression model fitted to the garnets' lattice constants in the dataset. The predicted values obtained as a result of the training of the model exhibited high correlation with the actual lattice constants, having the correlation coefficient of 0.988. For each composition in our dataset, the relative error between experimental and calculated lattice constants was less than 1.60%. %0 Journal Article %@ 0263-5747 %A La, Hung Manh %A Dinh, Tran Hiep %A Pham, Nhan Huu %A Ha, Quang Phuc %A Pham, Anh Quyen %D 2019 %F SisLab:3566 %J Robotica %N 5 %P 947-967 %T Automated robotic monitoring and inspection of steel structures and bridges %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3566/ %V 37 %0 Journal Article %@ 1932-6203 %A Lasko, Kristofer %A Vadrevu, Krishna Prasad %A Nguyen, Thi Nhat Thanh %D 2018 %F SisLab:3128 %J PLOS ONE %N 5 %T Analysis of air pollution over Hanoi, Vietnam using multi-satellite and MERRA reanalysis datasets %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3128/ %V 13 %X Air pollution is one of the major environmental concerns in Vietnam. In this study, we assess the current status of air pollution over Hanoi, Vietnam using multiple different satellite datasets and weather information, and assess the potential to capture rice residue burning emissions with satellite data in a cloud-covered region. We used a timeseries of Ozone Monitoring Instrument (OMI) Ultraviolet Aerosol Index (UVAI) satellite data to characterize absorbing aerosols related to biomass burning. We also tested a timeseries of 3-hourly MERRA-2 reanalysis Black Carbon (BC) concentration data for 5 years from 2012–2016 and explored pollution trends over time. We then used MODIS active fires, and synoptic wind patterns to attribute variability in Hanoi pollution to different sources. Because Hanoi is within the Red River Delta where rice residue burning is prominent, we explored trends to see if the residue burning signal is evident in the UVAI or BC data. Further, as the region experiences monsoon-influenced rainfall patterns, we adjusted the BC data based on daily rainfall amounts. Results indicated forest biomass burning from Northwest Vietnam and Laos impacts Hanoi air quality during the peak UVAI months of March and April. Whereas, during local rice residue burning months of June and October, no increase in UVAI is observed, with slight BC increase in October only. During the peak BC months of December and January, wind patterns indicated pollutant transport from southern China megacity areas. Results also indicated severe pollution episodes during December 2013 and January 2014. We observed significantly higher BC concentrations during nighttime than daytime with peaks generally between 2130 and 0030 local time. Our results highlight the need for better air pollution monitoring systems to capture episodic pollution events and their surface-level impacts, such as rice residue burning in cloud-prone regions in general and Hanoi, Vietnam in particular. %0 Journal Article %@ 1471-2105 %A Laukens, Kris %A Hollunder, Jens %A Dang, Thanh Hai %A De Jaeger, Geert %A Kuiper, Martin %A Witters, Erwin %A Verschoren, Alain %A Van Leemput, Koenraad %D 2010 %F SisLab:2078 %J BMC Bioinformatics %N 1 %P 360 %T Flexible network reconstruction from relational databases with Cytoscape and CytoSQL %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2078/ %V 11 %0 Conference Paper %A Le, Anh Cuong %A Nguyen, Tuan Phong %A Tran, Quoc Long %A Dao, Bao Linh %B 2018 10th International Conference on Knowledge and Systems Engineering (KSE) %C Ho Chi Minh city, Vietnam %D 2018 %F SisLab:3266 %P 79-84 %T Integrating Word Embeddings into IBM Word Alignment Models %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3266/ %X Wordalignmentmodelsareusedtogenerate word-aligned parallel text which is used in statistical machine translation systems. Currently, the most popular word alignment models are IBM models which have been widely applied in a large number of translation systems. The parameters of IBM models are estimated by using Maximum Likelihood principle, i.e. by counting the cooccurrence of words in the parallel text. This way of parameter estimation leads to the “ambiguity” problem when some words stand together in many sentence pairs but each of them is not translation of any other. Additionally, this method requires large amount of training data to achieve good results. However, parallel text which is used to train the IBM models is usually limited for low-resource languages. In this work, we try to solve these two problems by adding semantic information to the models. Our semantic information is derived from word embeddings which only need monolingual data to train. We deploy evaluation on a language pair that has great differences in grammar structure, English-Vietnamese. Even with this challenged task, our proposed models gain significant improvements in word alignment result and help increasing translation quality. %0 Conference Paper %A Le, Ba Cuong %A Nguyen, Van Son %A Nguyen, Duc Anh %A Pham, Ngoc Hung %A Vo, Dinh Hieu %B 4th International Conference on Information Systems Design and Intelligent Applications %C Danang, Vietnam %D 2017 %F SisLab:2467 %T JCIA: A Tool for Change Impact Analysis of Java EE Applications %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2467/ %X This paper presents a novel tool for change impact analysis of Java EE applications named JCIA. Analyzing the source code of the Java EE applications is a big challenge because of complexity and large scale of the applications. Moreover, components in Java EE applications are not only in Java language but also in different languages such as XHTML, XML, JSP, etc. This tool analyses source code of Java EE applications for building the dependency graphs (called JDG). The main idea for generating JDG is based on developing the source code analyzers for the typical technologies of Java EE such as Java Server Faces, Context and Dependency Injection, Web services, etc. Based on the obtained JDG and the given change sets, JCIA calculates the corresponding impact sets by applying the change impact analysis (CIA) based on change types and Wave-CIA method. The calculated impact sets help managers planning and estimating changes, developers implementing changes, and testers with regression testing. %0 Conference Paper %A Le, Ba Cuong %A Pham, Ngoc Hung %A Vo, Dinh Hieu %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1524 %T Một số phương pháp phân tích sự ảnh hưởng khi thay đổi mã nguồn phần mềm %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1524/ %X Bảo trì phần mềm hiện nay được đánh giá là một trong những giai đoạn khó khăn nhất, tốn kém nhất và đòi hỏi kỹ thuật cao trong suốt quá trình phát triển phần mềm. Các sản phẩm phần mềm thường xuyên có những thay đổi như thêm mới yêu cầu, sửa lỗi phần mềm trong khi lập trình viên thường chỉ có mã nguồn, không có hoặc thiếu tài liệu phát triển phần mềm. Mỗi sự thay đổi có thể tác động đến rất nhiều phần khác nhau của chương trình. Phân tích sự ảnh hưởng khi thay đổi phần mềm (Software change impact analysis - CIA) là các kỹ thuật nhằm xác định những tác động này, đóng vai trò quan trọng trong việc hiểu về chương trình, dự đoán những tác động, đánh giá chi phí, lựa chọn các ca kiểm thử để kiểm thử lại cũng như thực hiện chiến lược sửa đổi. Các kỹ thuật CIA được quan tâm của rất nhiều chuyên gia, từ tiếp cận theo hướng phân tích tĩnh cũng như phân tích động chương trình. Tuy nhiên những kết quả hiện tại chưa đủ mạnh để áp dụng trong thực tiễn khi độ chính xác của các kỹ thuật này còn thấp, tập các thành phần bị ảnh hưởng bởi thay đổi tìm được thường quá lớn. Bên cạnh đó các kỹ thuật CIA cho các ứng dụng phức tạp (khối lượng mã nguồn lớn, chứa nhiều loại ngôn ngữ lập trình, …) chưa thu được kết quả khả quan. Chính vì vậy chúng tôi dự kiến phát triển các kỹ thuật CIA đáp ứng những vấn đề trên với các mục tiêu cụ thể: • Đề xuất kỹ thuật CIA mới cải tiến độ chính xác của tập IS dựa trên sự kết hợp một số kỹ thuật truyền thống với các kỹ thuật dựa trên lịch sử phiên bản phần mềm • Nghiên cứu khả năng ứng dụng trong việc lựa chọn ca kiểm thử để kiểm thử lại và lựa chọn kế hoạch sửa đổi phần mềm khi có yêu cầu nâng cấp, dựa trên việc phân hạng các phần tử trong tập IS • Phát triển các công cụ hỗ trợ các phương pháp đề xuất và thực nghiệm với những phần mềm kích thước lớn, độ phức tạp cao Một số kết quả dự kiến: • 01 bài báo đăng trên tạp chí khoa học quốc tế. • 02 bài báo đăng trên tạp chí khoa học có uy tín ở trong nước. • 02 bài báo cáo trên hội nghị quốc tế. • Đề xuất phương pháp mới với độ chính xác cao hơn các phương pháp trước đây • Xây dựng công cụ hỗ trợ có thể ứng dụng trong thực tế • Đề xuất định hướng ứng dụng các kết quả của kỹ thuật CIA trong bảo trì, kiểm thử lại phần mềm. %0 Journal Article %@ 0866-8612 %A Le, Chi Luan %A Tran, Hoang Viet %A Pham, Ngoc Hung %D 2016 %F SisLab:1980 %I VNU Journal of Science %J VNU Journal of Computer Science and Communication Engineering %N 2 %P 31-42 %T A Framework for Modeling and Modular Verifying of Component-Based System Designs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1980/ %V 32 %X This paper introduces a framework for modeling and verifying safety properties of component-based systems (CBS) by extracting their models from designs in form of UML 2.0 sequence diagrams. Given UML 2.0 sequence diagrams of CBS, the framework extracts regular expressions exactly describing behaviors of the system. From these expressions, the proposed framework then generates accurate models represented by labeled transition systems (LTSs). After that, these models are used to modular check whether given designs satisfy required safety properties by using the assume-guarantee reasoning paradigm. This framework is not only useful for modeling and verifying designs at design phase, but also for effectively rechecking CBS in the context of software evolution. Implemented tools and experimental results are also presented in order to show the feasibilities and effectiveness of the proposed framework. %0 Conference Paper %A Le, Chi Luan %A Tran, Hoang Viet %A Pham, Ngoc Hung %B The 9th Asian Conference on Intelligent Information and Database Systems %C Kanazawa, Japan %D 2017 %F SisLab:2699 %P 549-558 %T On Implementation of the Assumption Generation Method for Component-Based Software Verification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2699/ %X The assume-guarantee verification has been recognized as a promising method for solving the state space explosion in modular model checking of component-based software. However, the counterexample analysis technique used in this method has huge complexity and the computational cost for generating assumptions is very high. As a result, the method is difficult to be applied in practice. Therefore, this paper presents two improvements of the assume-guarantee verification method in order to solve the above problems. The first one is a counterexample analysis method that is simple to implement but effective enough to prevent the verification process from infinite loops when considering the last action of counterexample as suffix in implementation. This is done by finding a suffix that can make the observation table not closed when being added to the suffix set of the table and use that suffix for the learning process. The second one is a reduction of the number of membership queries to be asked to teacher when learning assumptions. This results in a significantly faster speed in generating assumption than that of the original algorithm. An implemented tool and experimental results are also described to show the effectiveness of the improvements. %0 Conference Paper %A Le, Dieu Thu %A Nguyen, Cam Tu %A Ha, Quang Thuy %A Phan, Xuan Hieu %A Horiguchi, Susumu %B The 2008 IEEE/WIC/ACM International Conference on Web Intelligence %C Sydney, Australia %D 2008 %F SisLab:2298 %P 888-891 %T Matching and Ranking with Hidden Topics towards Online Contextual Advertising %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2298/ %0 Conference Paper %A Le, Dinh Minh %A Vuong, Tung Long %A Hoang, Van Xiem %A Dinh, Trieu Duong %A Le, Thanh Ha %B the 2016 International Conference Advanced Technologies for Communications (ATC) %C Hanoi, Vietnam %D 2016 %F SisLab:1998 %P 1-6 %T Improving 3D-TV View Synthesis Using Motion Compensated Temporal Interpolation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1998/ %X Nowadays, the development of three-dimension (3D) video applications such as three-dimensional television (3D-TV) and free-viewpoint television (FTV) has greatly increased human experiences. View synthesis method like depth-image-based-rendering (DIBR), plays a significant role in 3D content creation, 3D transmission, and has been integrated into video coding standards such as 3D-High efficiency video coding (3D-HEVC). However, the current DIBR method employs only the disparity correlation between views to create a so-called synthesized view; thus, unable to take full advantages of available synthesized information. In this paper, we propose a novel view synthesis method which takes advantages of not only the disparity correlation but also the temporal correlation between views. In the proposed method, an effective motion compensation based frame interpolation is employed to generate a temporal prediction view which is then combined with the DIBR rendered view to obtain the final synthesized view. Experimental results show that the proposed method can achieve the synthesized view with significantly outperforming other conventional techniques in terms of both peak signal-to noise ratio (PSNR) and subjective visual quality. %0 Journal Article %A Le, Dinh Thanh %D 2009 %F SisLab:1899 %J International Journal of Information Processing Systems %N 3 %P 145-150 %T Topological boundary detection in wireless sensor networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1899/ %V 5 %0 Conference Paper %A Le, Dinh Thanh %A Ho, Thuan %A Nguyen, Dai Tho %A VNU-UET, %B Hội nghị Khoa học về Công nghệ thông tin Trường Đại học Sư phạm Hà Nội %C Hanoi %D 2012 %F SisLab:131 %P 49-54 %T More Efficient Path Optimization for Greedy Geography Routing in Wireless Sensor Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/131/ %0 Conference Paper %A Le, Dinh Thanh %A Nguyen, Dai Tho %A University of Engineering and Technology, Vietnam National University, Hanoi, %B 2010 IEEE-RIVF International Conference on Computing and Communication Technologies %C Hanoi, Vietnam %D 2010 %F SisLab:2377 %P 148-153 %T Greedy Geographic Routing with Path Optimization in Wireless Sensor Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2377/ %X In this paper, we propose Greedy with Path Optimization Routing (GPOR), a novel geographic routing protocol for wireless sensor networks. GPOR finds initial routing paths by following a greedy with recovery strategy, then uses a follow-up technique to optimize the paths. An attempt is also made to create routing entries applicable to destination areas rather than individual nodes. Main advantages of GPOR are path optimization and void avoidance capacities. We implement GPOR in ns-2 and present simulation results. %0 Conference Paper %A Le, Dinh Thanh %A Nguyen, Dai Tho %A Ho, Thuan %A University of Engineering and Technology, Vietnam National University, Hanoi, %A Institute of Information Technology, Vietnam Academy of Science and Technology, %B Second International Symposium on Information and Communication Technology (SoICT 2011) %C Hanoi, Vietnam %D 2011 %F SisLab:2375 %P 86-91 %T Hybrid Contention-Based Geographic Routing in Wireless Sensor Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2375/ %X Beaconless and contention-based geographic routing is an attractive approach to resource-constrained wireless sensor networks. Aggressive contention is the most cost-efficient form of contention-based geographic routing since it uses no control packet. Nevertheless, to avoid duplicated data packets, aggressive contention must set restriction on the contention area. Consequently, its packet delivery rate is limited. On the other hand, non-aggressive contention maximizes packet delivery rate by making use of full contention area. As the compensation, control packets have to be used and additional delay is introduced. In this paper, we propose Hybrid Contention-Based Geographic Routing (HCGR) - a protocol that takes full aggressive contention and uses non-aggressive contention for recovering aggressive contention from failure. If aggressive contention succeeds, non-aggressive contention is suppressed. In cases where aggressive contention fails, non-aggressive contention is taken place to deliver data packets. Thus, HCGR can maximize packet delivery rate while keeping its overheads reasonably low. We implement HCGR and single-form protocols in the network simulator ns-2, conduct extensive simulations and present simulation results. %0 Conference Paper %A Le, Dinh Thanh %A Phan, Xuan Tien %B The 9th International Conference on Knowledge and Systems Engineering (KSE) %C Hue, Vietnam %D 2017 %F SisLab:2639 %T On the usage of character distribution for the detection of web attacks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2639/ %X Character distribution has been extensively used in literature to build models for the detection of web attacks. This paper explores that character distribution models should be built at attribute level in order to achieve a reasonable accuracy. However, attaching detection models to every single attribute leads to high memory and time complexities, which make attribute-specific models less practical. To remove these barriers, a simple yet effective solution has been proposed. In more details, by exploiting the language function of characters, character distribution can be reduced in size and rearranged in an intentional manner so that both time and memory complexities are reduced significantly. Detection models that use minimized and rearranged character distribution are, therefore, highly efficient and practical, especially suitable to large, high-traffic web applications. %0 Report %9 Technical Report %A Le, Duc-Trong %D 2019 %F SisLab:3724 %T Finding Relevant Files for Bug Reports Based on Mean Reciprocal Rank Maximization Approach %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3724/ %X Constructing and maintaining open source projects is not an easy work. Developers usually need to tackle a lot of bugs reported by users. The first step of the fixing progress is to find all relevant files respect to given reports. This step takes time and human resources. Motivating from this context, various learning-to-rank models were proposed in order to automatically rank files and generate suggestions for developers. The actual related files are expected to be appeared in high positions of the ranking. In the scope of this paper, a mean reciprocal rank optimization approach is investigated for the learning-to-rank relevant files for bug re- ports task. Given a bug report, the ranking of a source file is approximated by a function aggregating features which represent their relationship. The weights of these features are learned previously to maximize the mean reciprocal rank of known relevant files on training bug reports. In the ex- perimental section, the introduced model is evaluated on three Java open source projects namely Tomcat, AspectJ and SWT. The three different versions of the model are also explored and compared to a recent state-of-the-art method in recommending related files for bug reports. %0 Conference Paper %A Le, Duc-Trong %A Lauw, Hady W. %A Fang, Yuan %B the Twenty-Sixth International Joint Conference on Artificial Intelligence %C Melbourne, Australia %D 2017 %F SisLab:3709 %T Basket-Sensitive Personalized Item Recommendation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3709/ %X Personalized item recommendation is useful in narrowing down the list of options provided to a user. In this paper, we address the problem scenario where the user is currently holding a basket of items, and the task is to recommend an item to be added to the basket. Here, we assume that items currently in a basket share some association based on an underlying latent need, e.g., ingredients to prepare some dish, spare parts of some device. Thus, it is important that a recommended item is relevant not only to the user, but also to the existing items in the basket. Towards this goal, we propose two approaches. First, we explore a factorization-based model called BFM that incorporates various types of associations involving the user, the target item to be recommended, and the items currently in the basket. Second, based on our observation that various recommendations towards constructing the same basket should have similar likelihoods, we propose another model called CBFM that further incorporates basket-level constraints. Experiments on three real-life datasets from different domains empirically validate these models against baselines based on matrix factorization and association rules. %0 Conference Paper %A Le, Duc-Trong %A Lauw, Hady W. %A Fang, Yuan %B the Twenty-Eighth International Joint Conference on Artificial Intelligence %C Macau, China %D 2019 %F SisLab:3707 %T Correlation-Sensitive Next-Basket Recommendation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3707/ %X Items adopted by a user over time are indicative of the underlying preferences. We are concerned with learning such preferences from observed sequences of adoptions for recommendation. As multiple items are commonly adopted concurrently, e.g., a basket of grocery items or a sitting of media consumption, we deal with a sequence of baskets as input, and seek to recommend the next basket. Intuitively, a basket tends to contain groups of related items that support particular needs. Instead of recommending items independently for the next basket, we hypothesize that incorporating information on pairwise correlations among items would help to arrive at more coherent basket recommendations. Towards this objective, we develop a hierarchical network architecture codenamed Beacon to model basket sequences. Each basket is encoded taking into account the relative importance of items and correlations among item pairs. This encoding is utilized to infer sequential associations along the basket sequence. Extensive experiments on three public real-life datasets showcase the effectiveness of our approach for the next-basket recommendation problem. %0 Conference Paper %A Le, Duc-Trong %A Lauw, Hady W. %A Fang, Yuan %B the Twenty-Seventh International Joint Conference on Artificial Intelligence (IJCAI-18) %C Stockholm, Sweden %D 2018 %F SisLab:3708 %T Modeling Contemporaneous Basket Sequences with Twin Networks for Next-Item Recommendation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3708/ %X Our interactions with an application frequently leave a heterogeneous and contemporaneous trail of actions and adoptions (e.g., clicks, bookmarks, purchases). Given a sequence of a particular type (e.g., purchases)-- referred to as the target sequence, we seek to predict the next item expected to appear beyond this sequence. This task is known as next-item recommendation. We hypothesize two means for improvement. First, within each time step, a user may interact with multiple items (a basket), with potential latent associations among them. Second, predicting the next item in the target sequence may be helped by also learning from another supporting sequence (e.g., clicks). We develop three twin network structures modeling the generation of both target and support basket sequences. One based on "Siamese networks" facilitates full sharing of parameters between the two sequence types. The other two based on "fraternal networks" facilitate partial sharing of parameters. Experiments on real-world datasets show significant improvements upon baselines relying on one sequence type. %0 Conference Paper %A Le, Duc-Trong %A Lauw, Hady W. %A Fang, Yuan %B European Conference on Machine Learning and Principles and Practice of Knowledge Discovery %C Rivar De Garda, Italy %D 2016 %F SisLab:3710 %T Modeling sequential preferences with dynamic user and context factors %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3710/ %X Users express their preferences for items in diverse forms, through their liking for items, as well as through the sequence in which they consume items. The latter, referred to as “sequential preference”, manifests itself in scenarios such as song or video playlists, topics one reads or writes about in social media, etc. The current approach to modeling sequential preferences relies primarily on the sequence information, i.e., which item follows another item. However, there are other important factors, due to either the user or the context, which may dynamically a↵ect the way a sequence unfolds. In this work, we develop generative modeling of sequences, incorporating dynamic user-biased emission and context-biased transition for sequential preference. Experiments on publicly-available real-life datasets as well as synthetic data show significant improvements in accuracy at predicting the next item in a sequence. %0 Conference Paper %A Le, Hoang Quynh %A Can, Duy Cat %A Dang, Thanh Hai %A Tran, Mai Vu %A Ha, Quang Thuy %A Collier, Nigel %B The 9th International Conference on Knowledge and Systems Engineering (KSE) %C Hue, Vietnam %D 2017 %F SisLab:2707 %P 300-305 %T Improving Chemical-induced Disease Relation Extraction with Learned Features Based on Convolutional Neural Network %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2707/ %X There have been an increasing number of various machine learning-based models successfully proposed and applied for automatic chemical-induced disease (CID) relation extraction. They, however, usually require carefully handcrafted rich feature sets, which rely on expert knowledge, thus require expensive human labor but normally still cannot generalize data well enough. In this paper, we propose a CID relation extraction model that learns features automatically through a Convolutional Neural Network (CNN) instead of traditional handcrafted features. We exploit the shortest dependency path between a disease and a chemical for identifying their CID relation. Dependency relations, with and without their direction information, are further investigated. Experimental results on benchmark datasets (namely the BioCreative V dataset) are very potential, demonstrating the effectiveness of our proposed model for CID relation extraction. %0 Journal Article %A Le, Hoang Quynh %A Can, Duy Cat %A Ha, Quang Thuy %A Collier, Nigel %D 2019 %F SisLab:3477 %I Association for Computational Linguistics %J 2019 Annual Conference of the North American Chapter of the Association for Computational Linguistics %P 2902-2912 %T A Richer-but-Smarter Shortest Dependency Path with Attentive Augmentation for Relation Extraction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3477/ %V 1 %X To extract the relationship between two entities in a sentence, two common approaches are (1) using their shortest dependency path (SDP) and (2) using an attention model to capture a context-based representation of the sentence. Each approach suffers from its own disadvantage of either missing or redundant information. In this work, we propose a novel model that combines the advantages of these two approaches. This is based on the basic information in the SDP enhanced with information selected by several attention mechanisms with kernel filters, namely RbSP (Richer-but- Smarter SDP). To exploit the representation behind the RbSP structure effectively, we develop a combined deep neural model with a LSTM network on word sequences and a CNN on RbSP. Experimental results on the SemEval-2010 dataset demonstrate improved performance over competitive baselines. The data and source code are available at https: //github.com/catcd/RbSP. %0 Conference Paper %A Le, Hoang Quynh %A Can, Duy Cat %A Vu, Tien Sinh %A Dang, Thanh Hai %A Pilehvar, Mohammad Taher %A Collier, Nigel %B Conference on Empirical Methods in Natural Language Processing (EMNLP 2018) %C Brussels, Belgium %D 2018 %F SisLab:3171 %P 2266-2277 %T Large-scale Exploration of Neural Relation Classification Architectures %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3171/ %X Experimental performance on the task of relation classification has generally improved using deep neural network architectures. One major drawback of reported studies is that individual models have been evaluated on a very narrow range of datasets, raising questions about the adaptability of the architectures, while making comparisons between approaches difficult. In this work, we present a systematic large-scale analysis of neural relation classification architectures on six benchmark datasets with widely varying characteristics. We propose a novel multi-channel LSTM model combined with a CNN that takes advantage of all currently popular linguistic and architectural features. Our ‘Man for All Seasons’ approach achieves state-of-the-art performance on two datasets. More importantly, in our view, the model allowed us to obtain direct insights into the continued challenges faced by neural language models on this task. Example data and source code are available at: https://github. com/aidantee/MASS. %0 Conference Paper %A Le, Hoang Quynh %A Can, Duy Cat %A Vu, Tien Sinh %A Dang, Thanh Hai %A Pilehvar, Mohammad Taher %A Collier, Nigel %B The 2018 Conference on Empirical Methods in Natural Language Processing %C Brussels, Belgium %D 2018 %F SisLab:3342 %T Large-scale Exploration of Neural Relation Classification Architectures %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3342/ %0 Conference Paper %A Le, Hoang Quynh %A Tran, Mai Vu %A Dang, Thanh Hai %A Collier, Nigel %B The 5th BioCreative Challenge Evaluation %C Spain %D 2015 %F SisLab:1725 %T The UET-CAM System in the BioCreAtIvE V CDR Task %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1725/ %X Hoang-Quynh Le, Mai-Vu Tran, Thanh-Hai Dang, and Nigel Collier %0 Journal Article %@ 1758-0463 %A Le, Hoang Quynh %A Tran, Mai Vu %A Dang, Thanh Hai %A Ha, Quang Thuy %A Collier, Nigel %D 2016 %F SisLab:2472 %J Database %P baw102 %T Sieve-based coreference resolution enhances semi-supervised learning model for chemical-induced disease relation extraction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2472/ %V 2016 %X The BioCreative V chemical-disease relation (CDR) track was proposed to accelerate progress of text mining in facilitating integrative understanding ofchemical substances, diseases and their relations. In this article, we describe an extension of the UET-CAM system for mining chemical-disease relations from text data, of which performance was ranked 4th among 18 participating corresponding systems by the BioCreative CDR track committee. In Disease Named Entity Recognition and Normalization (DNER) phase, our system employs joint learning with a perceptron-based named entity recognizer (NER) and a back-off model with Semantic Supervised Indexing (SSI) and Skip-gram for named entity normalization (NEN). Crucially, for solving the chemical-induced disease (CID) sub-task, we propose a pipeline that includes a coreference resolution module and a SVM intra-sentence relations extraction model. The former module utilizes a multi-pass sieve to identify inter-sentence references for entities while the latter is trained on both the CDR data and our silverCID corpus with a rich feature set. SilverCID is the silver standard corpus contains more than 50 thousands sentences which are automatically built based on the CTD database in order to provide evidence for the CID relation extraction. We critically evaluated our method on the CDR test set in order to clarify the contribution of our system components. Results show an F1 of 82.44 for the DNER task, and a best performance of F1 58.90 on the CID task. The comparisons also demonstrate the significant contribution of the multi-pass sieve coreference resolution method and the silverCID corpus. %0 Conference Paper %A Le, Hoang Quynh %A Tran, Mai Vu %A Dang, Thanh Hai %A Ha, Quang Thuy %A Nigel, Collier %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1525 %T Sieve-based coreference resolution enhances semi-supervised learning model for chemical-induced disease relations extraction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1525/ %X The BioCreative V chemical-disease relation (CDR) track was proposed to accelerate progress of text mining in facilitating integrative understanding of chemical substances, diseases and their relations. In this article, we describe an extension of the UET-CAM system for mining chemical- disease relations from text data, of which performance was ranked 4th among 18 participating corresponding systems by the BioCreative CDR track committee. In Disease Named Entity Recognition and Normalization (DNER) phase, our system employs joint learning with a perceptron-based named entity recognizer (NER) and a back-off model with Semantic Supervised Indexing (SSI) and Skip-gram for named entity normalization (NEN). Crucially, for solving the chemical-induced disease (CID) sub-task, we propose a pipeline that includes a coreference resolution module and a SVM intra-sentence relations extraction model. The former module utilizes a multi-pass sieve to identify inter-sentence references for entities while the latter is trained on both the CDR data and our silverCID corpus with a rich feature set. SilverCID is the silver standard corpus contains more than 50 thousands sentences which are automatically built based on the CTD database in order to provide evidence for the CID relation extraction. We critically evaluated our method on the CDR test set in order to clarify the contribution of our system components. Results show an F1 of 82.44 for the DNER task, and a best performance of F1 58.90 on the CID task. The comparisons also demonstrate the significant contribution of the multi-pass sieve coreference resolution method and the silverCID corpus. %0 Conference Paper %A Le, Hoang Quynh %A Tran, Mai Vu %A Dang, Thanh Hai %A Nigel, Collier %B 2015: the 5th BioCreative challenge evaluation workshop %C Sevilla, Spain %D 2015 %F SisLab:1610 %P 208-214 %T The UET-CAM System in the BioCreAtIvE V CDR Task %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1610/ %X In this paper, we describe a system developed for the BioCreative Vchemical-disease relation (CDR) task. The Disease Named Entity Recognition and Normalization (DNER) model employs joint learning using a perceptronbased named entity recognizer (NER) and a back-off model for named entity normalization (NEN). In order to maximize both precision and recall, our NEN adopts a sequential back-off ensemble approach based on Semantic Supervised Indexing (SSI) - a supervised Word Embedding (WE) method- giving results by inference from training data, and Skip-gram - an unsupervised WE methodtaking advantage of large unlabeled data. In the Chemical-induced diseases relation extraction (CID) model, we firstly resolve co-references by using a multipass sieve to identify cross-sentence references for entities, thus enabling intrasentence relations to be discovered more easily. Following this we extract CID relations using a support vector machine model trained on supervised sentence data from the CDR training and development dataset. We evaluated our method on both the DNER test set and the CID test set. Results show an F1 of 76.44 for the DNER tas %0 Journal Article %@ 1383-469X %A Le, Hong Anh %A Dao, Thi Huong %A Truong, Ninh Thuan %D 2017 %F SisLab:2481 %I Springer %J Mobile networks and Applications %N 2 %P 356-366 %T A formal approach to checking consistency in software refactoring %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2481/ %V 22 %X In software development, refactoring is a process that improves the system internal structure without altering its external behavior. Applying design patterns, which are common reusable solutions of several kinds of problems is widely adopted. This technique, however, raises a challenging issue that after applying design patterns the software system may not preserve some certain behavioral properties. This paper proposes a new approach to checking consistency between original software system and its evolution at both design and implementation phases. First, we formalize elements of software designs and programs. Methods, based on these formalizations, are proposed for verifying the design and implementation of the system. Finally, the paper presents a case study of Adaptive Road Traffic Control system to illustrate the proposed approach in detail. %0 Journal Article %A Le, Hong Anh %A Shin, Nakajima %A Truong, Ninh Thuan %D 2016 %F SisLab:1900 %I Springer %J SpringerPlus %N 1000 %T Formal analysis of imprecise system requirements with Event‑B %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1900/ %V 5 %X Formal analysis of functional properties of system requirements needs precise descriptions. However, the stakeholders sometimes describe the system with ambiguous, vague or fuzzy terms, hence formal frameworks for modeling and verifying such requirements are desirable. The Fuzzy If–Then rules have been used for imprecise requirements representation, but verifying their functional properties still needs new methods. In this paper, we propose a refinement-based modeling approach for specification and verification of such requirements. First, we introduce a representation of imprecise requirements in the set theory. Then we make use of Event-B refinement providing a set of translation rules from Fuzzy If–Then rules to Event-B notations. After that, we show how to verify both safety and eventuality properties with RODIN/EventB. Finally, we illustrate the proposed method on the example of Crane Controller. %0 Conference Paper %A Le, Hong Anh %A Truong, Ninh Thuan %B 2014: the 2nd International Conference on Context-Aware Systems and Applications (ICCASA) %C Phu Quoc, Vietnam %D 2014 %F SisLab:1296 %P 250-259 %T Formal Modeling and Verification of Context-Aware Systems using Event-B %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1296/ %X Context awareness is a computing paradigm that makes applications responsive and adaptive with their environment. Formal modeling and verification of context-aware systems are challenging issues in the development as they are complex and uncertain. In this paper, we propose an approach to use a formal method Event-B to model and verify such systems. First, we specify a context aware system’s components such as context data entities, context rules, context relations by Event-B notions. In the next step, we use the Rodin platform to verify the system’s desired properties such as safety properties. It aims to benefit from natural representation of context awareness concepts in Event-B and proof obligations generated by refinement mechanism to ensure the correctness of systems. We illustrate the use of our approach on a simple example. %0 Conference Paper %A Le, Hong Anh %A Truong, Ninh Thuan %B 5th EAI International Conference on Context-Aware Systems and Applications (ICCASA) %C Viet Nam %D 2016 %F SisLab:1901 %T A New Method to Analyze Graphical User Interfaces of Android Applications %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1901/ %X In recent years, the number of Android smartphones increase dramatically and new applications are added numerously in Google store. Android developers usually have to deal with the difficulties such as limited capacity battery, screen design, and limited resources. Among them, specifying graphical user interfaces (GUI) of an application is one of the most important issues. This paper presents a new method to analyze GUI specifications of an Android application. We employ Event-B formal method and its refinement mechanism to formalize the specifications and to check if the constraints are satisfied. A running example of a Note application is given to illustrate the proposed method in detail. %0 Conference Paper %A Le, Hong Anh %A Truong, Ninh Thuan %A Nakajima, Shin %B SAC 2015 %D 2015 %F SisLab:1647 %P 1651-1653 %T Verifying eventuality properties of imprecise system requirements using event-B %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1647/ %0 Conference Paper %A Le, Hong Hai %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1526 %T Fingerprint Matching on GPU for a Large Scale Identification System %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1526/ %X Fingerprints are most used biometrics features for identification. Although state-of-the-art algorithms are very accurate, but the need for processing speed for databases with millions fingerprints are very demanding. GPU devices are used widely in parallel computing tasks for their efficiency and low-cost. Most approaches make use of GPU for the filtering process in a multi-stage matching system. In this paper, we present a complete fingerprint matching algorithm on GPU. Our approach uses minutia cylinder-code (MCC) representation with a global consolidation stage and a careful design to make it suitable for the architecture of GPU. The result tested with GTX- 680 device shows that the proposed algorithm can perform 1.8 millions matches per second, making it applicable for real time identification systems with databases of millions fingerprints. %0 Conference Paper %A Le, Hong Hai %A Nguyen, Ngoc Hoa %B The 7th International Symposium on Information and Communication Technology (SoICT) %C Ho Chi Minh city, Vietnam %D 2016 %F SisLab:2250 %T Nâng cao thuật toán đối sánh mã trụ MCC %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2250/ %X Vân tay là đặc trưng sinh trắc học được sử dụng phổ biến rộng rãi trong các ứng dụng dân sự cũng như pháp lý. Hiện nay các kết quả nghiên cứu đạt được các kết quả tương đối tốt về độ chính xác và tốc độ xử lý, nhưng nhu cầu về các hệ thống nhận dạng nhanh, chính xác, đặc biệt trên các CSDL lớn hàng triệu vân tay vẫn là bài toán nhận được nhiều quan tâm nghiên cứu hiện nay. Trong bài báo này, chúng tôi khảo sát tổng quan về một số thuật toán đối sánh vân tay hiện tại, chú trọng đến phương pháp đối sánh dựa trên mã trục các điểm trạc MCC. Từ đó, chúng tôi đã đề xuất kỹ thuật cải tiến MCC dựa trên bước gia cố. Kỹ thuật này đã được tiến hành thực nghiệm trên bộ CSDL vân tay chuẩn FVC 2002. Kết quả đánh giá chúng tôi thu được có cả ba độ đo EER, FMR100, FMR1000 chính xác hơn so với giải thuật MCC chuẩn trong khi tốc độ so khớp gần như được giữ nguyên. %0 Conference Paper %A Le, Hong Hai %A Nguyen, Ngoc Hoa %B The 2013 IEEE RIVF International Conference on Computing and Communication Technologies (RIVF) %C Hanoi, Vietnam %D 2013 %F SisLab:2251 %T Toward an approach to improving the consolidation stage of fingerprint matching %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2251/ %X Consolidation stage is a component of fingerprint matching algorithms. It aims to check whether the local matching minutiae pairs still satisfy at the global alignment level of two fingerprints. In this paper, we review some consolidation methods used in the consolidation stage and propose a verification step to improve the quality of the consolidation methods by removing unmatched minutiae pairs. We carry out our proposal on two popular fingerprint matching algorithms: K-plet based and m-Triplet based one. The results of our experiments on FVC 2002-DB1 database show that the modified algorithms get better accuracy and not affect to the speed of the original one. %0 Conference Paper %A Le, Hong Hai %A Nguyen, Ngoc Hoa %A Nguyen, Tri Thanh %B The 9th Asian Conference on Intelligent Information and Database Systems %C Kanazawa, Japan %D 2017 %F SisLab:2323 %T Automatic Detection of Singular Points in Fingerprint Images using Convolution Neural Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2323/ %X Minutiae based matching, the most popular approach used in fingerprint matching algorithms, is to calculate the similarity by finding the maximum number of matched minutiae pairs in two given fingerprints. With no prior knowledge about anchor/clue to match, this becomes a combinatorial problem. Global features of the fingerprints (e.g., singular core and delta points) can be used as the anchor to speed up the matching process. Most approaches using the conventional Poincare Index method with additional techniques to improve the detection of the core and delta points. Our approach uses Convolution Neural Networks which gained state-of-the-art results in many computer vision tasks to automatically detect those points. With the experimental results on FVC2002 database, we achieved the accuracy and false alarm of (96%, 7.5%) and (90%, 6%) for detecting core, and delta points, correspondingly. These results are comparative to those of the detection algorithms with human knowledge. %0 Conference Paper %A Le, Hong Hai %A Nguyen, Ngoc Hoa %A Nguyen, Tri Thanh %B 8th Asian Conference on Intelligent Information and Database Systems (ACIIDS) %C Danang, Vietnam %D 2016 %F SisLab:1871 %I Springer %P 688-697 %T Exploiting GPU for Large Scale Fingerprint Identification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1871/ %0 Journal Article %@ 2475-1839 %A Le, Hong Hai %A Nguyen, Ngoc Hoa %A Nguyen, Tri Thanh %D 2018 %F SisLab:3361 %I Taylor & Francis %J Journal of Information and Telecommunication %P 1-16 %T Speeding up and enhancing a large-scale fingerprint identification system on GPU %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3361/ %X Fingerprint identification is one of the most common biometric feature problems which is used in many applications. Although state-of-the-art algorithms are very accurate, the need for fast processing a big database of millions of fingerprints is highly demanding. In this paper, we propose to adapt the fingerprint matching algorithm based on minutia cylinder-code (MCC) on Graphics Processing Units (GPUs) to speed up the matching. Another contribution of this paper is to add a consolidation stage after matching to enhance the precision. The experimental results on a GTX-680 and a K40 tesla devices with standard datasets prove the proposed algorithm can be comparable with the state-of-the-art approach, and suitable for a real time identification application. %0 Conference Paper %A Le, Hong Hai %A Nguyen, Ngoc Hoa %A Nguyen, Tri Thanh %B the 7th International Conference on Information Science and Application (ICISA) %C Ho Chi Minh city, Vietnam %D 2016 %F SisLab:1869 %I Springer %P 679-688 %T A complete fingerprint matching algorithm on GPU for a large scale identification system %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1869/ %0 Conference Paper %A Le, Hong Phuong %A Phan, Xuan Hieu %A Nguyen, Tien Dung %B The 8th International Conference on Knowledge and Systems Engineering (KSE) %C Hanoi, Vietnam %D 2014 %F SisLab:1885 %P 653-665 %T Using Dependency Analysis to Improve Question Classification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1885/ %V 326 %X Question classification is a first necessary task of automatic question answering systems. Linguistic features play an important role in developing an accurate question classifier. This paper proposes to use typed dependencies which are extracted automatically from dependency parses of questions to improve accuracy of classification. Experiment results show that with only surface typed dependencies, one can improve the accuracy of a discriminative question classifier by over 8.0% on two benchmark datasets. %0 Conference Paper %A Le, Hong Phuong %A Phan, Xuan Hieu %A Tran, The Trung %B The 2013 IEEE RIVF International Conference on Computing and Communication Technologies, Research, Innovation, and Vision for the Future (RIVF) %C Hanoi, Vietnam %D 2013 %F SisLab:1886 %P 103-108 %T On the effect of the label bias problem in part-of-speech tagging %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1886/ %X This paper investigates the effect of the label bias problem of maximum entropy Markov models for part-of-speech tagging, a typical sequence prediction task in natural language processing. This problem has been underexploited and underappreciated. The investigation reveals useful information about the entropy of local transition probability distributions of the tagging model which enables us to exploit and quantify the label bias effect of part-of-speech tagging. Experiments on a Vietnamese treebank and on a French treebank show a significant effect of the label bias problem in both of the languages. %0 Conference Paper %A Le, Khanh Trinh %B The 12th International Symposium on Theoretical Aspects of Software Engineering %C Guangzhou, China %D 2018 %F SisLab:2965 %P 1-8 %T Formalization and Verification of AUTOSAR OS Standard's Memory Protection %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2965/ %X AUTOSAR OS is a standard for automotive operating systems, which provides a specification that consists ofu nctionalities such as scheduling services, timing services, and memory protection. In this paper, we focus on memory protection features among them. As the AUTOSAR OS specification is described in natural language, its ambiguity may confuse developers as well as cause the contradiction of the specification, then eventually lead to serious problems of automotive systems such as bugs and errors. These problems in automotive systems relate directly to the safety of human being. Thus, it is very important to ensure the unambiguity and consistency of the specification. Our solution for the problems is formalizing the AUTOSAR OS specification using Event-B specification language which allows us to formally specify the functionalities of AUTOSAR OS and reduce the ambiguity of natural language. We developed a formal specification of the memory protection of AUTOSAR OS and verified its consistency. In this verification, we found the inconsistency of the specification during discharging proof obligations generated by RODIN which is a tool for Event-B. This inconsistency comes from the ambiguity of the original specification, and finding it by reviewing based on natural language description is very hard. In this paper, we explain how we found the inconsistency existed in the AUTOSAR OS standard after showing our approach to formalize and verify it with Event-B. %0 Journal Article %A Le, Khanh Trinh %A Vo, Dinh Hieu %A Pham, Ngoc Hung %D 2015 %F SisLab:1743 %J Research, Development and Application on Information Technology and Communication Technology %N 8 %P 28-37 %T A Method for Automated User Interaction Testing of Web Applications %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1743/ %V E3 %0 Conference Paper %A Le, Kim Thu %A Dang, Cao Cuong %A Le, Sy Vinh %B 2018 10th International Conference on Knowledge and Systems Engineering (KSE) (KSE'18) %D 2018 %F SisLab:3279 %P 242-246 %T Building a specific amino acid substitution model for dengue viruses %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3279/ %X Phylogenetic trees inferred from protein sequences are strongly affected by amino acid substitution models. Although different amino acid substitution models have been proposed, only a few were estimated for specific species such as the FLU model for influenza viruses. Among the most dangerous viruses for human health, dengue is always on top and the cause of dengue fever up to 100 million people per year. In this study, we built a specific amino acid substitution model for dengue protein sequences, called DEN. The dengue protein sequences were obtained from the NCBI dengue database and the model was estimated using the maximum likelihood method. Experiments showed that the new model DEN helped to build better phylogenetic trees than other existing models. We strongly recommend researchers to use the DEN model for analyzing dengue protein data. %0 Conference Paper %A Le, Minh Duc %A Dang, Duc Hanh %A Nguyen, Viet Ha %B 12th International Conference on Computing and Communication Technologies (RIVF) %C Hanoi, Vietnam %D 2016 %F SisLab:1950 %I IEEE %P 247-252 %T Domain-Driven Design Patterns: A Metadata-Based Approach %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1950/ %0 Conference Paper %A Le, Minh Duc %A Dang, Duc Hanh %A Nguyen, Viet Ha %B The 8th International Conference on Knowledge and Systems Engineering (KSE) %C Hanoi, Vietnam %D 2016 %F SisLab:1951 %I IEEE %P to appear %T Domain-Driven Design Using Meta-Attributes: A DSL-Based Approach %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1951/ %0 Journal Article %@ 0950-5849 %A Le, Minh Duc %A Dang, Duc Hanh %A Nguyen, Viet Ha %D 2020 %F SisLab:4146 %I Elsevier %J Information and Software Technology %P 1-32 %T Generative Software Module Development for Domain-Driven Design with Annotation-Based Domain Specific Language %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4146/ %V 120 %0 Conference Paper %A Le, Minh Duc %A Dang, Duc Hanh %A Nguyen, Viet Ha %B The 9th International Conference on Knowledge and Systems Engineering (KSE) %C Hue, Vietnam %D 2017 %F SisLab:2671 %P 1-7 %T Generative Software Module Development: A Domain-Driven Design Perspective %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2671/ %0 Journal Article %@ 1477-8424 %A Le, Minh Duc %A Dang, Duc Hanh %A Nguyen, Viet Ha %D 2018 %F SisLab:3140 %I Elsevier %J Computer Languages, Systems and Structures %P 199-235 %T On Domain Driven Design Using Annotation-Based Domain Specific Language %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3140/ %V 54 %X The aim of object-oriented domain-driven design (DDD) is to iteratively develop software around a realistic domain model. Recent work in DDD use an annotation-based extension of object-oriented programming language to build the domain model. This model becomes the basis for a ubiquitous language and is used as input to generate software. However, the annotation-based extensions of these work do not adequately address the primitive and essential structural and behavioural modelling requirements of practical software. Further, they do not precisely characterise the software that is generated from the domain model. In this paper, we propose a DSL-based DDD method to address these limitations. We make four contributions: (1) An annotation-based domain-specific language (DSL) named DCSL, whose annotation extension expresses a set of essential structural constraints and the essential behaviour of a domain class. (2) A structural mapping between the state and behaviour spaces of a domain class. This mapping enables a technique for generating the behavioural specification. (3) A technique that uses DCSL to support behavioural modelling with UML activity diagram. (4) A 4-property characterisation of the software generated from the domain model. We demonstrate our method with a Java software tool and evaluate DCSL in the context of DDD. %0 Conference Paper %A Le, Minh Duc %A Dang, Duc Hanh %A Vu, Thanh Ha %B SoICT 2019: 10th International Symposium on Information and Communications %C Ha Long Bay %D 2019 %F SisLab:3581 %T jDomainApp: A Module-Based Domain-Driven Software Framework %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3581/ %X Object-oriented domain-driven design (DDD) has been advocated to be the most common form of DDD, thanks to the popularity of object-oriented development methodologies and languages. Although the DDD method prescribes a set of design patterns for the domain model, it provides no languages or tools that realize these patterns. There have been several software frameworks developed to address this gap. However, these frameworks have not tackled two important software construction issues: generative, module-based software construction and development environment integration. In this paper, we propose a framework, named jDomainApp, and an Eclipse IDE plugin to address these issues. In particular, we extend our recent works on DDD to propose a software configuration language that expresses the software configuration, needed to automatically generate software from a set of modules. The modules are automatically generated using a module configuration language that we defined in a previous work. We demonstrate the framework and plug-in using a real-world software example. Further, we evaluate the performance of software construction to show that it is scalable to handle large software. %0 Conference Paper %A Le, Minh Duc %A Nguyen, Hoa Huy %A Nguyen, Duc Loc %A Nguyen, Viet Anh %B Proceedings of the 2020 The 6th International Conference on Frontiers of Educational Technologies %C Tokyo Japan %D 2020 %F SisLab:4059 %T How to Forecast the Students' Learning Outcomes Based on Factors of Interactive Activities in a Blended Learning Course %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4059/ %X This paper summarizes the research results of identifying the influencing factors in the online learning phase of a blended learning course. From such factors, we propose a model for predicting student outcomes. In our study, we have conducted several models in order to predict the student's learning outcomes, using a course of 231 participants. Obtained data from the logs file of an LMS system is analyzed using learning analytics and machine learning techniques, and the results propose that the four factors are the number of views, the number of posts, the number of forum views, and the number of on-time submitted assignments impact on the student's learning outcomes. For the forecast of the final exam grade based on the results of the formative assessment tests, Bayesian Ridge is the most accurate among the four conducted models (Linear Regression, KNR, SVM, Bayesian Ridge). Our study can be a useful material for lecturers and course designers in effectively organizing blended learning courses. %0 Conference Paper %A Le, Minh Khoi %A Chang, Hsien-Tsung %A Chang, Yi-Min %A Hu, Yi-Hao %A Chen, Huan-Ting %B 2016 International Computer Symposium %C Chiayi, Taiwan %D 2016 %F SisLab:2059 %K cloud database; smart clothes; Spark; %T An efficient multilevel healthy cloud system using Spark for smart clothes %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2059/ %X Nowadays, we can find part of people into the suboptimal health status. However, most of them have not enough time for health examinations regularly. Smart clothes which are wearable and convenient devices for monitoring physiological signals solve this problem. In this study, we develop an efficient multilevel healthy cloud system to analyzing data collected from CGU smart clothes. The cloud system classifies the raw data into three types: normal data, continuous data, and multimedia data. After that, the cloud system analyzes data and predict diseases using Apache Spark. And then, the system will transfer final results to smartphones. The experimental result shows that the cloud system achieves very high-speed performance. Moreover, the system has friendly health information presentation and effective authorization process to help doctors and patients contact each other. %0 Journal Article %@ 0866-8612 %A Le, Nguyen Khoi %A Landa-Silva, Dario %D 2016 %F SisLab:2330 %I VNU Journal of Science %J VNU Journal of Computer Science and Communication Engineering %N 1 %P 10-32 %T Hyper-volume Evolutionary Algorithm %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2330/ %V 32 %X We propose a multi-objective evolutionary algorithm (MOEA), named the Hyper-volume Evolutionary Algorithm (HVEA). The algorithm is characterised by three components. First, individual fitness evaluation depends on the current Pareto front, specifically on the ratio of its dominated hyper-volume to the current Pareto front hyper-volume, hence giving an indication of how close the individual is to the current Pareto front. Second, a ranking strategy classifies individuals based on their fitness instead of Pareto dominance, individuals within the same rank are non guaranteed to be mutually non-dominated. Third, a crowding assignment mechanism that adapts according to the individual's neighbouring area, controlled by the neighbouring area radius parameter, and the archive of non-dominated solutions. We perform extensive experiments on the multiple 0/1 knapsack problem using different greedy repair methods to compare the performance of HVEA to other MOEAs including NSGA2, SEAMO2, SPEA2, IBEA and MOEA/D. This paper shows that by tuning the neighbouring area radius parameter, the performance of the proposed HVEA can be pushed towards better convergence, diversity or coverage and this could be beneficial to different types of problems. %0 Book Section %A Le, Nguyen Tuan Thanh %A Pham, Manh Linh %B Industry 4.0 Interoperability, Analytics, Security, and Case studies %C Boca Raton, FL, USA %D 2021 %E Rajesh, G. %E X. Mercilin, Raajini %E Dang, Thi Thu Hien %F SisLab:4374 %I CRC Press - Taylor & Francis Group, LLC %K Big Data Analytics, Industry 4.0, Machine Learning, Deep Learning %P 1-11 %S Big Data for Industry 4.0: Challenges and Applications %T Big Data Analytics and Machine Learning for Industry 4.0: An Overview %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4374/ %X The concept of “Big data” was mentioned for the first time by Roger Mougalas in 2005. Volume hints to the size and/or scale of datasets. Until now, there is no universal threshold for data volume to be considered as big data, because of the time and diversity of datasets. Velocity indicates the speed of processing data. It can fall into three categories: streaming processing, real-time processing, or batch processing. Value alludes to the usefulness of data for decision making. Veracity denotes the quality and trustworthiness of datasets. Parallelization allows one to improve computation time by dividing big problems into smaller instances, distributing smaller tasks across multiple threads and then performing them simultaneously. Feature selection is useful for preparing high scale datasets. Sampling is a method for data reducing that helps to derive patterns in big datasets by generating, manipulating, and analyzing subsets of the original data. %0 Conference Paper %A Le, Pham Van Linh %A Tran, Quang Bach %A Pham, Tien Lam %A Tran, Quoc Long %B KNOWLEDGE AND SYSTEMS ENGINEERING (KSE 2020) %D 2020 %F SisLab:4220 %T Node-aware convolution in Graph Neural Networks for Predicting molecular properties %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4220/ %X Molecular property prediction is a challenging task which aims to solve various issues of science namely drug discovery, materials discovery. It focuses on understanding the structure-property relationship between atoms in a molecule. Previous approaches have to face difficulties dealing with the various structure of the molecule as well as heavy computational time. Our model, in particular, utilizes the idea of message passing neural network and Schnet on the molecular graph with enhancement by adding the Node-aware Convolution and Edge Update layer in order to acquire the local information of the graph and to propagate interaction between atoms. Through experiments, our model has been shown the outperformance with previous deep learning methods in predicting quantum mechanical, calculated molecular properties in the QM9 dataset and magnetic interaction of two atoms in molecules approaches. %0 Conference Paper %A Le, Phe Do %B Hội thảo lần thứ II Một số vấn đề chọn lọc về An toàn, an ninh thông tin %D 2017 %F SisLab:2751 %T Các phương pháp tấn công hệ mã dòng xác thực hạng nhẹ ACORN–128 và đề xuất cải tiến để tăng độ an toàn %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2751/ %X Hệ mã xác thực nhẹ ACORN là các hệ mã xác thực tốt, tuy nhiên nó vẫn có những điểm yếu nhất định. Trong báo cáo này, chúng tôi tổng hợp một số phương pháp tấn công lên hệ ACORN: phương pháp tấn công khối của Itai Dinur và Adi Shamir và các cuộc tấn công phục hồi trạng thái. Tấn công khối là một dạng tấn công đại số cho phép đối phương hạ bậc của phương trình đa thức từ nguyên thủy mật mã. Các chuyên gia Md Iftekhar Salam, Harry Bartlett, Ed Dawson, Josef Pieprzyk, Leonie Simpson, and Kenneth Koon-Ho Wong đã sử dụng tấn công khối để rút ngắn vòng của ACORN. Tấn công khối trên 477 vòng khởi tạo của ACORN có thể khôi phục 128 bit khóa với độ phức tạp khoảng 235. Các chuyên gia chỉ ra, hệ phương trình tuyến tính liên quan đến trạng thái ban đầu của phiên bản đầy đủ của ACORN có thể dễ dàng tạo ra để thực hiện cuộc tấn công khôi phục trạng thái với độ phức tạp 227:8. Tấn công phục hồi trạng thái dựa vào tính chất trượt được các chuyên gia Meicheng và Dongdai Lin nghiên cứu. Các chuyên gia chỉ ra rằng, đối với mỗi cặp (Key, IV) có cặp khác cùng tạo ra một trạng thái tại các thời điểm khác nhau với xác suất 1. Để khắc phục các nhược điểm của hệ ACORN – 128 hiện nay, chúng tôi đề xuất thay các thanh ghi dịch tuyến tính bởi các thanh ghi dịch phi tuyến, tăng độ dài khóa, độ dài véc tơ khởi tạo và mỗi khóa, mỗi véc tơ khởi tạo chỉ sử dụng một lần. Các phương pháp này được dẫn chứng và minh họa cụ thể. Hy vọng, những thông tin đưa lại bổ ích cho các chuyên gia lập mã để tạo ra hệ mã xác thực nhẹ tôt hơn để sử dụng hiệu quả trong IoT. Keywords-Mã dòng, Mã dòng nhẹ, Mã dòng nhẹ xác thực, ACORN, Phương pháp tấn công khối, Phương pháp tấn công dựa vào tính chất trượt, ... %0 Conference Paper %A Le, Phe Do %B Kỷ yếu Hội thảo lần I: Một số vấn đề chọn lọc về an toàn, an ninh thông tin %C Hanoi, Vietnam %D 2016 %F SisLab:2058 %T Trường Đại học Công nghệ-ĐHQGHN, ngôi trường tiên phong trong nghiên cứu và đào tạo về an toàn thông tin %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2058/ %X Các giảng viên của Trường ĐHCN - ĐHQG HN bắt đầu nghiên cứu và đào tạo về An toàn thông tin từ rất sớm với những cây đại thụ như GS. TSKH Phan Đình Diệu, PGS. TS Trịnh Nhật Tiến. Trong thời gian qua, trường ĐHCN - ĐHQG HN đã có nhiều đề tài Luận án Tiến sĩ, Luận văn Thạc sĩ và Khóa luận tốt nghiệp của sinh viên về An tòn thông tin. ĐHQG HN đã cho phép Trường ĐHCN mở chuyên ngành đào tạo về An toàn thông tin. Côn g tác nghiên cứu và đào tạo về ATTT ở Trường ĐHCN - ĐHQG HN ngày càng phát triển, ngày càng có sự kết nối giữa CNTT và ĐT - VT và các khoa khác trong trường, các đơn vị khác trong và ngoài nước. %0 Conference Paper %A Le, Phe Do %A Le, Trung Thuc %A Mai, Manh Trung %A Le, Thi Len %A Nguyen, Khac Hung %A Nguyen, Thi Hang %A Tran, Van Manh %B Hội thảo Quốc gia lần thứ XX: Một số vấn đề chọn lọc của Công nghệ thông tin và truyền thông - Quy Nhơn, 23 - 24/11/2017 %D 2017 %F SisLab:2749 %T Cải tiến mã khối hạng nhẹ họ LED và Neokeon %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2749/ %X Họ mã khối Neokeon được đề xuất vào năm 2000 bởi bốn nhà khoa học Daemen, Peeters, Van Assche và Rijme. Môt trong các ưu điểm của Neokeon đó là nhỏ gọn, nhanh chóng triển khai trên phần cứng chuyên dụng, ngoài ra nó còn yêu cầu về bộ nhớ RAM rất thấp và hoạt động hiệu quả trên đa nền tảng. Nhưng trong thuật toán họ Neokeon sử dụng rất nhiều các ánh xạ tuyến tính và sử dụng khóa trực tiếp trong khi mã hóa mà không thông qua những phép biến đổi làm việc trước với khóa. Do đó, Neokeon dường như khá mong manh trước những dạng thám mã tuyến tính. LED là thuật toán được thiết kế và đề xuất bởi Guo, Peyrin, Poschmann và Robshaw vào năm 2011, là một trong những thuật toán mã hóa nhẹ khá mới. Kích thước khối tin 64-bit với hai biến thể của khóa lần lượt là 64-bit và 128-bit. Thiết kế của LED có nhiều điểm tương đồng với thuật toán AES. Trong bài báo này, chúng tôi thực hiện cải tiến thuật toán LED và Neokeon, đối với Neokeon chúng tôi ngẫu nhiên hóa dữ liệu tại những ánh xạ tuyến tính và thao tác làm việc với khóa, nhằm mục đích nâng cao độ an toàn của Neokeon trước các thám mã tuyến tính nhưng không làm thay đổi nhiều những ưu điểm sẵn có của Neokeon, đối với LED chúng tôi thực hiện cải tiến xử lý dữ liệu trong khối step của thuật toán này. Keywords-mã khối hạng nhẹ; led; neokeon; tuyến tính %0 Conference Paper %A Le, Phe Do %A Le, Trung Truc %A Mai, Manh Trung %A Le, Thi Thu Thao %A LêLe, Thi Len %A Do, Dang Thuan %A Nguyen, Van Thang %A Do, Cong Thanh %B Kỷ yếu Hội thảo Quốc gia lần thứ 17 "Một số vấn đề chọn lọc của CNTT & TT" %C Dac Lac %D 2015 %F SisLab:1618 %P 88-94 %T Hệ mật mã kép an toàn %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1618/ %0 Journal Article %@ 1859-1043 %A Le, Phe Do %A Mai, Manh Trung %A Le, Trung Truc %A Nguyen, Thi Hang %A Vuong, Thi Hanh %A Nguyen, Khac Hung %A Dinh, Thi Thuy %A Le, Thi Len %D 2017 %F SisLab:2757 %J Tạp chí Nghiên cứu Khoa học & Công nghệ Quân sự %T Nghiên cứu một số hệ mật mã nhẹ và ứng dụng trong IoT %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2757/ %X Người ta ước tính đến năm 2020 sẽ có hơn 50 tỷ thiết bị kết nối internet, nghĩa là mỗi người trên trái đất trung bình sẽ có 6,6 đồ vật trực tuyến. Trái đất sẽ được che phủ bởi hàng triệu cảm biến thu thập thông tin và tải lên internet. Để đảm bảo các kết nối được an ninh và an toàn các thiết bị đó cần có các hệ mật vừa có độ mật cần thiết, tiêu tốn ít năng lượng, bộ nhớ và các cổng logic. Đó là các hệ mật mã nhẹ, gồm mã khối hạng nhẹ, mã dòng hạng nhẹ và các mã xác thực hạng nhẹ. Trong báo cáo này chúng tôi giới thiệu một số hệ mật trong mật mã nhẹ, đưa ra những điểm mạnh và điểm yếu của chúng. Các hệ mã khối hạng nhẹ được nghiên cứu gồm Klein, Led, Present, Mini – AES, Mcrypyon và Katan. Hệ mã dòng được chúng tôi giới thiệu là Grain. Kết quả có thể dùng làm tài liệu tham khảo cho các nhà chuyên môn về mật mã nhẹ và IoT. %0 Journal Article %@ 1859-1043 %A Le, Phe Do %A Mai, Manh Trung %A Le, Trung Truc %A Nguyen, Thi Hang %A Vuong, Thi Hanh %A Nguyen, Khac Hung %A Dinh, Thi Thuy %A Le, Thi Len %D 2017 %F SisLab:2551 %J Tạp chí Nghiên cứu Khoa học & Công nghệ Quân sự %P 134-147 %T Nghiên cứu một số hệ mật mã nhẹ và ứng dụng vào IoT %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2551/ %V 5 %X Người ta ước tính đến năm 2020 sẽ có hơn 50 tỷ thiết bị kết nối internet, nghĩa là mỗi người trên trái đất trung bình sẽ có 6,6 đồ vật trực tuyến. Trái đất sẽ được che phủ bởi hàng triệu cảm biến thu thập thông tin và tải lên internet. Để đảm bảo các kết nối được an ninh và an toàn các thiết bị đó cần có các hệ mật vừa có độ mật cần thiết, tiêu tốn ít năng lượng, bộ nhớ và các cổng logic. Đó là các hệ mật mã nhẹ, gồm mã khối hạng nhẹ, mã dòng hạng nhẹ và các mã xác thực hạng nhẹ. Trong báo cáo này chúng tôi giới thiệu một số hệ mật trong mật mã nhẹ, đưa ra những điểm mạnh và điểm yếu của chúng. Các hệ mã khối hạng nhẹ được nghiên cứu gồm Klein, Led, Present, Mini – AES, Mcrypyon và Katan. Hệ mã dòng được chúng tôi giới thiệu là Grain. Kết quả có thể dùng làm tài liệu tham khảo cho các nhà chuyên môn về mật mã nhẹ và IoT. %0 Conference Paper %A Le, Phe Do %A Nguyen, Canh Hoang %B Hội thảo Quốc gia lần thứ 18 "Một số vấn đề chọn lọc của CNTT & TT" %C Ho Chi Minh city %D 2015 %F SisLab:1620 %T Entropy và ngưỡng an toàn trong lưu trữ và truyền thông tin %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1620/ %0 Conference Paper %A Le, Phe Do %A Nguyen, Canh Hoang %B Hội nghị toàn quốc lần thứ 4 về Xác suất - thống kê %C Danang, Vietnam %D 2015 %F SisLab:1619 %T Phân tích hồi quy và đánh giá chất lượng cuộc sống %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1619/ %0 Conference Paper %A Le, Quang Dao %A Nguyen, Nhat Nam %A Luu, Quang Thang %A Ha, Duc Van %A Ha, Minh Cuong %A Pham, Bao Son %A Bui, Quang Hung %B Hội thảo GIS toàn quốc 2020 %D 2020 %F SisLab:4101 %I Trường Đại học Bách Khoa – ĐHQG-HCM %T Research and application of D-LINKNET network to solve the problem of detecting road from WorldView-3 satellite image in Cau Giay district, Hanoi %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4101/ %X In this paper, we have outlined the need for solving the road detection problem, from which to compare and find the type of data to be used as Worldview-3 satellite images, to find out the method of success. Very good achievement in the DeepGlobe contest is D-LinkNet. This thesis has collected Worldview-3 and yandex image data for preprocessing data. We have collected and pre-processed Worldview-3 data, deployed and installed D-LinkNet. Application of evaluation method of IoU point of DeepGlobe. In addition, learn how to develop a realistic assessment method rather than a long-distance evaluation. D-LinkNet works well in sparsely populated areas, big roads such as suburbs or new urban areas, resettlement areas. In densely populated urban areas, small alleys are obscured by houses and trees, making it difficult and difficult to detect roads. Some areas with the same spectrum as the road are also mislabeled. %0 Conference Paper %A Le, Quang Hung %A Le, Anh Cuong %A Huynh, Van Nam %B 2013: %C Hanoi, VIETNAM %D 2013 %F SisLab:755 %P 175-179 %T Parallel Phrase Extraction from English-Vienamese Parllel Corpora %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/755/ %0 Conference Paper %A Le, Quang Hung %A Nguyen, Duy Cuong %A Pham, Duc Hong %A Le, Anh Cuong %B 2013: The Fifth International Conference on Knowledge and Systems Engineering (KSE 2013) %C Hanoi, Vietnam %D 2013 %F SisLab:754 %P 251-260 %T Paragraph Alignment for English-Vietnamese Parallel E-books %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/754/ %V 2 %0 Journal Article %A Le, Quoc Khanh %A Nguyen, Sy Tan %A Nguyen, Thi Nhat Thanh %A Le, Thanh Ha %D 2013 %F SisLab:786 %J IEIE Transactions on Smart Processing & Computing %N 3 %P 117-129 %T Burned Area Detection After Wildfire Using Landsat 7 ETM+ SLC-off Images %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/786/ %V 2 %0 Journal Article %A Le, Quoc Khanh %A Pham, Chinh Huu %A Le, Thanh Ha %D 2012 %F SisLab:1923 %J IEIE Transactions on Smart Processing & Computing %N 1 %P 1-7 %T Road traffic control gesture recognition using depth images %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1923/ %V 1 %0 Journal Article %@ 1537-1719 %A Le, Si Quang %A Dang, Cao Cuong %A Gascuel, Olivier %D 2012 %F SisLab:69 %I Oxford Journals %J Molecular Biology and Evolution %K amino acid substitutions %N 10 %P 2921-2936 %T Modeling Protein Evolution with Several Amino Acid Replacement Matrices Depending on Site Rates %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/69/ %V 29 %X Most protein substitution models use a single amino acid replacement matrix summarizing the biochemical properties of amino acids. However, site evolution is highly heterogeneous and depends on many factors that influence the substitution patterns. In this paper, we investigate the use of different substitution matrices for different site evolutionary rates. Indeed, the variability of evolutionary rates corresponds to one of the most apparent heterogeneity factors among sites, and there is no reason to assume that the substitution patterns remain identical regardless of the evolutionary rate. We first introduce LG4M, which is composed of four matrices, each corresponding to one discrete gamma rate category (of four). These matrices differ in their amino acid equilibrium distributions and in their exchangeabilities, contrary to the standard gamma model where only the global rate differs from one category to another. Next, we present LG4X, which also uses four different matrices, but leaves aside the gamma distribution and follows a distribution-free scheme for the site rates. All these matrices are estimated from a very large alignment database, and our two models are tested using a large sample of independent alignments. Detailed analysis of resulting matrices and models shows the complexity of amino acid substitutions and the advantage of flexible models such as LG4M and LG4X. Both significantly outperform single-matrix models, providing gains of dozens to hundreds of log-likelihood units for most data sets. LG4X obtains substantial gains compared with LG4M, thanks to its distribution-free scheme for site rates. Since LG4M and LG4X display such advantages but require the same memory space and have comparable running times to standard models, we believe that LG4M and LG4X are relevant alternatives to single replacement matrices. Our models, data, and software are available from http://www.atgc-montpellier.fr/models/lg4x. %0 Journal Article %A Le, Sy Vinh %A Dang, Cuong %A Le, Quang %D 2017 %F SisLab:2519 %J BMC Mol Evol %N 136 %T Improved mitochondrial amino acid substitution models for metazoan evolutionary studies %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2519/ %V 17 %X Background: Amino acid substitution models play an essential role in inferring phylogenies from mitochondrial protein data. However, only few empirical models have been estimated from restricted mitochondrial protein data of a hundred species. The existing models are unlikely to represent appropriately the amino acid substitutions from hundred thousands metazoan mitochondrial protein sequences. Results: We selected 125,935 mitochondrial protein sequences from 34,448 species in the metazoan kingdom to estimate new amino acid substitution models targeting metazoa, vertebrates and invertebrate groups. The new models help to find significantly better likelihood phylogenies in comparison with the existing models. We noted remarkable distances from phylogenies with the existing models to the maximum likelihood phylogenies that indicate a considerable number of incorrect bipartitions in phylogenies with the existing models. Finally, we used the new models and mitochondrial protein data to certify that Testudines, Aves, and Crocodylia form one separated clade within amniotes. Conclusions: We introduced new mitochondrial amino acid substitution models for metazoan mitochondrial proteins. The new models outperform the existing models in inferring phylogenies from metazoan mitochondrial protein data. We strongly recommend researchers to use the new models in analysing metazoan mitochondrial protein data. %0 Journal Article %@ 1552-4833 %A Le, Sy Vinh %A Le, Truc %A Le, Van Khanh %A Kieu, Huynh %A Do, Hang %D 2017 %F SisLab:2518 %I Wiley %J American Journal of Medical Genetics Part A %P 1-6 %T A mutation in GABRB3 associated with Dravet syndrome %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2518/ %V 9999 %X Dravet syndrome is a rare and severe type of epilepsy in infants. Approximately, 70–80% of patients with Dravet syndrome have mutations in SCN1A, the gene encoding the alpha-1 subunit of the sodium channel, while some simplex patients have variants in one of several other genes, including but not limited to GABRA1, SCN2A, STXBP1, GABRG2, and SCN1B. In this study, we performed exome sequencing in six patients with SCN1A-negative Dravet syndrome to identify other genes related to this disorder. In one affected individual, we detected a novel de novo heterozygous missense variant, c.695G>A, p.(Arg232Gln), in GABRB3, the gene encoding the β3-subunit of the gamma-aminobutyric acid type A (GABAA) receptor, which mediates inhibitory signaling within the central nervous system. In summary, the data in this study identify GABRB3 as a candidate gene for Dravet syndrome. %0 Conference Paper %A Le, Sy Vinh %A Nguyen, Duc Canh %A Bui, Ngoc Thang %A Do, Thi Thu Hang %A Duong, Quoc Chinh %A Tran, Cong Hoang %A Le, Ba Hong Minh %A Pham, Thi Dieu Linh %B The 9th International Conference on Knowledge and Systems Engineering (KSE) %C Hue, Vietnam %D 2017 %F SisLab:2688 %I IEEE %P 142-147 %T Genomedics: Whole exome analysis system for clinical studies %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2688/ %X Whole exome sequencing (WES) is a widely used technique in both medical studies and clinical practice. However, a number of studies show that the results produced by different WES analysis pipelines are not always homogeneous. To this end, we propose a method (called Genomedics) using a consensus approach to expand the list of variants by combining results called from six separate pipelines with sensitive options. To evaluate the performance of the proposed method, Gemomedics was compared to seven existing methods when they were tested on two datasets and F1-score was used as an indicator of accuracy. The results showed that Genomedics has the highest score among seven methods. We also applied Genomedics to analyze whole exomes from Multiple Myeloma and Dravet syndrome patients and found interesting results. The results demonstrate the promising applications of Genomedics in clinical studies. %0 Patent %@ 1-2020-01334 %A Le, Thanh Ha %A Bui, The Duy %A Ngo, Thi Duyen %A Nguyen, Van Tung %A Nguyen, The Hoang Anh %A Nguyen, Minh Hoa %D 2020 %F SisLab:4222 %T PHƯƠNG PHÁP VÀ HỆ THỐNG NHẬP LIỆU DỰA TRÊN BÀN PHÍM ẢO VÀ CHUYỂN THÀNH ÂM THANH %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4222/ %0 Conference Paper %A Le, Thanh Ha %A Cho, Jae-Young %A Kim, Hyun-Ji %A Ko, Sung-Jea %B 2013: International Conference on Electronics, Information and Communication (ICEIC) %D 2013 %F SisLab:789 %T Ilumination-Adaptive Color Reproduction Method for Daylight Condition %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/789/ %0 Journal Article %@ 1225-6463 %A Le, Thanh Ha %A Jung, Seung Won %A Baek, Seung Jin %A Ko, Sung Jea %D 2011 %F SisLab:1905 %J ETRI Journal %N 2 %P 194-200 %T Improved Method for the Macroblock-Level Deblocking Scheme %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1905/ %V 33 %0 Journal Article %A Le, Thanh Ha %A Jung, Seung Won %A Choi, Kang Sun %A Ko, Sung Jea %D 2010 %F SisLab:1918 %I IET %J Electronics letters %N 16 %P 1121-1123 %T Image segmentation based on modified graph-cut algorithm %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1918/ %V 46 %0 Conference Paper %A Le, Thanh Ha %A Jung, Seung-Won %A Bui, The Duy %A Dinh, Trieu Duong %A Ko, Sung-Jea %B International Conferences, CA and CES3 2011 %C Jeju Island, Korea %D 2011 %F SisLab:1929 %I Springer %P 357-363 %T Motion vector refinement algorithm for the high definition video coding %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1929/ %0 Journal Article %A Le, Thanh Ha %A Jung, Seung-Won %A Park, Chun-Su %A Ko, Sung-Jea %D 2010 %F SisLab:1913 %I Electronics and Telecommunications Research Institute %J ETRI journal %N 2 %P 336-338 %T Macroblock-level deblocking method to improve coding efficiency for H. 264/AVC %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1913/ %V 32 %0 Conference Paper %A Le, Thanh Ha %A Nguyen, Hai Chau %A Nguyen, Nam Hoang %A Bui, Quang Hung %A Nguyen, Thi Nhat Thanh %A Nguyen, Thanh Thuy %A Nguyen, Ba Tung %A Do, Khac Phong %A Nguyen, Van Duong %A Phan, Van Trong %A Pham, Thanh Tung %A Nguyen, Danh Thanh Hai %A Vuong, Van Quynh %A Pham, Van Cu %B BESETOHA %C Hanoi, Vietnam %D 2014 %F SisLab:1421 %T A Review of Forest Fire Information Technologies in Vietnam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1421/ %0 Journal Article %@ 0269-7491 %A Le, Thanh Ha %A Nguyen, Thi Nhat Thanh %A Lasko, Kristofer %A Ilavajhala, Shriram %A Vadrevu, Krishna Prasad %A Justice, Chris %D 2014 %F SisLab:1419 %J Environmental Pollution %N 75 %P 195-267 %T Vegetation Fires and Air Pollution in Vietnam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1419/ %X Forest fires are a significant source of air pollution in Asia. In this study, we integrate satellite remote sensing data and ground-based measurements to infer fire-air pollution relationships in selected regions of Vietnam. We first characterized the active fires and burnt areas at a regional scale from MODIS satellite data. We then used satellite-derived active fire data to correlate the resulting atmospheric pollution. Further, we analyzed the relationship between satellite atmospheric variables and ground-based air pollutant parameters. Our results show peak fire activity during March in Vietnam, with hotspots in the Northwest and Central Highlands. Active fires were significantly correlated with UV Aerosol Index (UVAI), aerosol extinction absorption optical depth (AAOD), and Carbon Monoxide. The use of satellite aerosol optical thickness improved the prediction of Particulate Matter (PM) concentration significantly. %0 Patent %@ 25669 %A Le, Thanh Ha %A Nguyen, Thị Thủy %A Tran, Quoc Long %A Nguyen, Do Van %A Nguyen, Chi Thanh %A Nguyen, Viet Anh %A Nguyen, Cong Thanh %D 2020 %F SisLab:4177 %T Phương pháp và thiết bị phân tích nội dung viđeo để phát hiện sự kiện người ngã %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4177/ %0 Journal Article %@ 1876-1100 %A Le, Thanh Ha %A Seung, Won Jung %A Chee, Sun Won %D 2015 %F SisLab:1599 %J Advanced Multimedia and Ubiquitous Engineering, Lecture Notes in Electrical Engineering %P 117-122 %T Reduced Reference Quality Metric for Depth Images %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1599/ %V Volume %X In this paper, a new quality metric for depth images is proposed. Unlike the conventional depth metrics which require the additional information such as the ground truth depth image or a stereo image pair, the proposed quality metric demands only a single camera image and its corresponding depth image. In this work, we first empirically observe that the depth distortion is closely related to the local image characteristics. Based on the observation, we introduce a method to assess the local depth distortion for the edge and non-edge regions. Then, the local distortion is adaptively weighted by the Gabor filter and added up to the quality metric for the depth image. %0 Journal Article %@ 2588-1086 %A Le, Thanh Xuyen %A Le, Trung Thanh %A Dinh, Van Viet %A Tran, Quoc Long %A Nguyen, Linh Trung %A Nguyen, Duc Thuan %D 2018 %F SisLab:4085 %I VNU %J VNU Journal of Science: Computer Science and Communication Engineering %N 2 %P 1-13 %T Deep Learning for Epileptic Spike Detection %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4085/ %V 33 %X In the clinical diagnosis of epilepsy using electroencephalogram (EEG) data, an accurate automatic epileptic spikes detection system is highly useful and meaningful in that the conventional manual process is not only very tedious and time-consuming, but also subjective since it depends on the knowledge and experience of the doctors. In this paper, motivated by significant advantages and lots of achieved successes of deep learning in data mining, we apply Deep Belief Network (DBN), which is one of the breakthrough models laid the foundation for deep learning, to detect epileptic spikes in EEG data. It is really useful in practice because the promising quality evaluation of the spike detection system is higher than $90$\%. In particular, to construct accurate detection model for non-spikes and spikes, a new set of detailed features of epileptic spikes is proposed. These features were then fed to the DBN which is modified from a generative model into a discriminative model to aim at classification accuracy. The experiment results indicate that it is possible to use deep learning models for epileptic spike detection with very high performance in item of sensitivity, selectivity, specificity and accuracy 92.82%, 97.83% , 96.41%, and 96.87%, respectively. %0 Conference Paper %A Le, Thanh Xuyen %A Le, Trung Thanh %A Nguyen, Linh Trung %A Tran, Thi Thuy Quynh %A Nguyen, Duc Thuan %B 2019 KICS Korea-Vietnam International Joint Workshop on Communications and Information Sciences %C Hanoi, Vietnam %D 2019 %F SisLab:3700 %T EEG tensor representation using graph signal processing %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3700/ %X To help analyze electroencephalography (EEG) data for epilepsy, we propose a temporal–spatial–spectral tensor tensor representation for epileptic spikes via graph signal processing. Instead of using the temporal behavior of the sources, we apply the graph wavelet transform to the spatial variable for constructing the new representation of the epilepsy tensor, in order to exploit latent information of the spatial domain. To illustrate the usefulness of the proposed method for EEG tensor representation, we decompose this tensor to separate EEG sources. Simulation results on real EEG data of a patient diagnosed of epilepsy indicated that our proposed method of representation is promising in estimating the epileptic sources. %0 Conference Paper %A Le, Thu %A Le, Sy Vinh %B RIVF 2020 %D 2020 %F SisLab:4223 %T A protein alignment partitioning method for protein phylogenetic inference %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4223/ %0 Conference Paper %A Le, Thu %A Le, Sy Vinh %A Do Duc, Dong %A Bui, Thang %A Nguyen, Thao %B KSE 2020 %C Can Tho %D 2020 %F SisLab:4225 %P 300-305 %T iK-means: an improvement of the iterative k-means partitioning algorithm %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4225/ %0 Conference Paper %A Le, Thu Hong %A Nguyen, Chi Thanh %A Tran, Quoc Long %B 2020 RIVF International Conference on Computing and Communication Technologies (RIVF) %D 2020 %F SisLab:4217 %T Polyp Segmentation in Colonoscopy Images Using Ensembles of U-Nets with EfficientNet and Asymmetric Similarity Loss Function %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4217/ %X Automatic polyp detection and segmentation are highly desirable for colon screening due to polyp miss rate by physicians during colonoscopy, which is about 25%. Diagnosis of polyps in colonoscopy videos is a challenging task due to variations in the size and shape of polyps. In this paper, we adapt U-net and evaluate its performance with different modern convolutional neural networks as its encoder for polyp segmentation. One of the major challenges in training networks for polyp segmentation raises when the data are unbalanced, polyp pixels are often much lower in numbers than non-polyp pixels. A trained network with unbalanced data may make predictions with high precision and low recall, being severely biased toward the non-polyp class which is particularly undesired because false negatives are more important than false positives. We propose an asymmetric similarity loss function to address this problem and achieve a much better tradeoff between precision and recall. Finally, we propose an ensemble method for further performance improvement. We evaluate the performance of well-known polyp datasets CVC-ColonDB and ETIS-Larib PolypDB. The best results are 89.13% dice, 79.77% IOU, 90.15% recall, and 86.28% precision. Our proposed method outperforms the state-of-the-art polyp segmentation methods. %0 Conference Paper %A Le, Trung Thanh %A Nguyen, Linh Trung %A Nguyen, Viet Dung %A Abed-Meraim, Karim %B The 4th NAFOSTED Conference on Information and Computer Science (NICS) %C Hanoi, Vietnam %D 2017 %F SisLab:3034 %P 150-155 %T A New Windowed Graph Fourier Transform %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3034/ %X Many practical networks can be mathematically modeled as graphs. Graph signal processing (GSP), intersecting graph theory and computational harmonic analysis, can be used to analyze graph signals. Just as short-time Fourier transform (STFT) for time-frequency analysis in classical signal processing, we have windowed graph Fourier transform (WGFT) for vertex-frequency analysis in GSP. In this paper, we introduced a new graph modulation operator that satisfies the property of spectral conservation, and a new graph translation operator with interesting properties. Based on these operators, we presented a new method to obtain the WGFT with a tight vertex-frequency frame. These GSP tools were developed based on the graph adjacency matrix. Using time-series graph, USA graph and random graph as examples, we showed by simulation the advantages of our proposed GSP tools over the state-of-the-arts. %0 Report %9 Technical Report %@ UET-AVITECH-2019001 %A Le, Trung Thanh %A Nguyen, Linh Trung %A Nguyen, Viet Dung %A Abed-Meraim, Karim %C Hanoi, Vietnam %D 2019 %F SisLab:3443 %T A new windowed digraph Fourier transform %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3443/ %X Signal processing on graphs is an emerging topic of signal processing, considering graphs as signals. Hence, similar concepts and tools in signal processing can be extended, for understanding graph signals, such as graph frequency, graph operators, graph Fourier transform. In this report, we focus on the joint relationship of vertex-frequency representation of graphs. Motivated by the existing relationship between the incidence matrix and the Laplacian matrix for undirected graphs, we define a generalized Laplacian matrix for directed graphs. This allows us to propose a unified framework for the graph Fourier transform, graph modulation and graph translation. Then, built upon the graph modulation and translation operators, we propose a new windowed graph Fourier transform which a tight vertex-frequency frame. The experimental results indicate that our proposed methods are better than state-of-the-art methods and beneficial for analyzing various different types of graphs, including directed, undirected, weighted, unweighted and connected graphs. %Z UET-AVITECH-2019001 %0 Report %9 Technical Report %@ UET-AVITECH-2019002 %A Le, Trung Thanh %A Nguyen, Thi Anh Dao %A Nguyen, Viet Dung %A Nguyen, Linh Trung %A Abed-Meraim, Karim %C Hanoi, Vietnam %D 2019 %F SisLab:3444 %T Simultaneous tensor decomposition for EEG epileptic spike detection %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3444/ %X Objective: Epilepsy is one of the most common brain disorders. For epilepsy diagnosis or treatment, the neurologist needs to observe epileptic spikes from electroencephalography (EEG) data. Since multi-channel EEG records can be naturally represented by multi-way tensors, it is of interest to see whether tensor decomposition is able to analyze EEG epileptic spikes. Approach: In this report, we first proposed the problem of simultaneous multilinear low-rank approximation of tensors (SMLRAT) and proved that SMLRAT can obtain local optimum solutions by using two well-known tensor decomposition algorithms (HOSVD and Tucker-ALS). Second, we presented a new system for automatic epileptic spike detection based on SMLRAT. Main results: We compared the proposed tensor analysis method with other common tensor methods in analyzing EEG signal and compared the proposed feature extraction method with Phan’s method. Experimental results indicated that our proposed method is able to detect epileptic spikes with good performance. Significance: To suitably deal with EEG spikes, we developed a local solution for nonnegative SMLRAT. For practical implementation, we proposed the generalized SMLRAT algorithm to effectively solve the SMLRAT and nonnegative SMLRAT problems. An efficient EEG feature extraction framework was proposed, based on estimating the “eigenspikes” from the nonnegative generalized SMLRAT algorithm. %Z UET-AVITECH-2019002 %0 Conference Paper %A Le, Trung Thanh %A Nguyen, Viet Dung %A Nguyen, Linh Trung %A Karim, Abed Meraim %B 27th European Signal Processing Conference (EUSIPCO) %C Coruna, Spain %D 2019 %F SisLab:3703 %T Robust subspace tracking with missing data and outliers via ADMM %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3703/ %X Robust subspace tracking is crucial when dealing with data in the presence of both outliers and missing observations. In this paper, we propose a new algorithm, namely PETRELS-ADMM, to improve performance of subspace tracking in such scenarios. Outliers residing in the observed data are first detected in an efficient way and removed by the alternating direction method of multipliers (ADMM) solver. The underlying subspace is then updated by the algorithm of parallel estimation and tracking by recursive least squares (PETRELS) in which each row of the subspace matrix was estimated in parallel. Based on PETRELS-ADMM, we also derive an efficient way for robust matrix completion. Performance studies show the superiority of PETRELS-ADMM as compared to the state-ofthe-art algorithms. We also illustrate its effectiveness for the application of background-foreground separation. %0 Report %9 Technical Report %@ UET-AVITECH-2019003 %A Le, Trung Thanh %A Nguyen, Viet Dung %A Nguyen, Linh Trung %A Karim, Abed Meraim %C Vietnam %D 2019 %F SisLab:3702 %I VNU University of Engineering and Technology, Vietnam National University, Hanoi %T Robust subspace tracking: Novel algorithm and performance guarantee %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3702/ %X Subspace tracking, which is refered to online PCA, is a classical problem in signal processing with various applications in wireless communications, rada and image/video processing. Since outliers and missing data are ubiquitous and more common in big data regime, robust variants of subspace tracking (RST) are crucial. In this paper, we propose a novel algorithm, namely PETRELSADMM, to improve RST performance in such scenario. The proposed approach consists of two main stages, including outlier rejection and subspace estimation. In the first stage, alternating direction method of multipliers (ADMM) solver is used to detect outliers residing in the observed data in an efficient way. In the second stage, we propose a modification of the parallel estimation and tracking by recursive least squares (PETRELS) algorithm to update the underlying subspace. A theoretical convegence analysis is provided, i.e., we prove that PETRELS-ADMM can generate a sequence of subspace solutions converging to the optimum of its batch counterpart. Performance studies show the superiority of our algorithms as compared to the state-of-the-art algorithms on both synthesis data and real data. %Z UET-AVITECH-2019003 %0 Journal Article %@ 1741-2552 %A Le, Trung Thanh %A Nguyen Thi Anh, Dao %A Nguyen, Viet Dung %A Nguyen, Linh Trung %A Karim, Abed-Meraim %D 2020 %F SisLab:4004 %I IOP %J Journal of Neural Engineering %T Multi-channel EEG epileptic spike detection by a new method of tensor decomposition %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4004/ %X Objective. Epilepsy is one of the most common brain disorders. For epilepsy diagnosis or treatment, the neurologist needs to observe epileptic spikes from electroencephalography (EEG) data. Since multi-channel EEG records can be naturally represented by multi-way tensors, it is of interest to see whether tensor decomposition is able to analyze EEG epileptic spikes. Approach. In this paper, we first proposed the problem of simultaneous multilinear low-rank approximation of tensors (SMLRAT) and proved that SMLRAT can obtain local optimum solutions by using two well-known tensor decomposition algorithms (HOSVD and Tucker-ALS). Second, we presented a new system for automatic epileptic spike detection based on SMLRAT. Main results. We propose to formulate the problem of feature extraction from a set of EEG segments, represented by tensors, as the SMLRAT problem. Efficient EEG features were obtained, based on estimating the ‘eigenspikes’ derived from nonnegative GSMLRAT. We compared the proposed tensor analysis method with other common tensor methods in analyzing EEG signal and compared the proposed feature extraction method with the state-of-the-art methods. Experimental results indicated that our proposed method is able to detect epileptic spikes with high accuracy. Significance. Our method, for the first time, makes a step forward for automatic detection EEG epileptic spikes based on tensor decomposition. The method can provide a practical solution to distinguish epileptic spikes from artifacts in real-life EEG datasets. %0 Conference Paper %A Le, Van Dat %A Dang, Cao Cuong %A Le, Si Quang %A Le, Sy Vinh %B 2011 Third International Conference on Knowledge and Systems Engineering (KSE) %D 2011 %F SisLab:70 %K amino acid substitution models, biology computing, evolution (biological), evolutionary information, genetics, maximum likelihood approaches, maximum likelihood estimation, phylogenetics trees, protein phylogenetics analysis, protein sequence alignment, Proteins, trees (mathematics) %P 85 -91 %T A Fast and Efficient Method for Estimating Amino Acid Substitution Models %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/70/ %X Amino acid substitution models (matrices) play important role for protein phylogenetics analysis and protein sequence alignment. Different approaches have been proposed to estimate amino acid substitution matrices since the time of Day Hoff in 1972. Currently, maximum likelihood approaches have been widely used to estimate popular matrices such as WAG, LG, FLU, etc. Although maximum likelihood approaches result in high quality matrices, they are slow and not applicable to very large datasets. The most time consuming step in estimating matrices is building phylogenetics trees from protein alignments. In this paper, we propose new methods to overcome the obstacle by splitting large alignments into small ones which still contain enough evolutionary information for estimating matrices. Experiments with both Pfam and FLU datasets showed that proposed methods were about three to nine times faster than the best current method while the quality of estimated matrices are nearly the same. Thus, our methods will enable researchers to estimate matrices from very large datasets. %0 Book Section %@ 0302-9743 %A Le, Van Giap %A Nguyen, Huu Tung %A Lu, Dang Nhac %A Nguyen, Ngoc Hoa %B Computational Collective Intelligence %D 2016 %F SisLab:2042 %I Springer %P 367-378 %S Lecture Notes in Computer Science %T A Solution for Automatically Malicious Web Shell and Web Application Vulnerability Detection %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2042/ %V 9875 %0 Journal Article %@ 2190-9288 %A Le, Van Giap %A Nguyen, Huu Tung %A Pham, Duy Phuc %A Nguyen, Ngoc Hoa %D 2018 %F SisLab:3375 %I Springer %J Transactions on Computational Collective Intelligence %N XXXII %P 182-208 %T GuruWS: A Hybrid Platform for Detecting Malicious Web Shells and Web Application Vulnerabilities %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3375/ %V 11370 %X Web application/service is now omnipresent but its security risks, such as malware and vulnerabilities, are indeed underestimated. In this paper, we propose a protective, extensible and hybrid platform, named GuruWS, for automatically detecting both web application vulnerabilities and malicious web shells. Based on the original PHP vulnerability scanner THAPS, we propose E-THAPS which implements a novel detection mechanism, an improved SQL injection, Cross-site Scripting and vulnerability detection capabilities. For malicious web shell detection, taint analysis and pattern matching methods are chosen to be implemented in GuruWS. A number of extensive experiments are carried out to prove the outstanding performance of our proposed platform in comparison with several existing solutions in detecting either web application vulnerabilities or malicious web shells. %0 Journal Article %@ 1738-7906 %A Le, Viet Ha %A Phung, Van On %A Nguyen, Ngoc Hoa %D 2020 %F SisLab:3991 %J IJCSNS International Journal of Computer Science and Network Security %N 6 %P 72-82 %T Information Security Risk Management by a Holistic Approach: a Case Study for Vietnamese e-Government %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3991/ %V 20 %X Information security risk management is one of the essential tasks currently in ensuring information security. In particular, for e-Government information systems, the assessment and management of security risks through the exploitation of software vulnerabilities, network equipment, etc., allow us to minimize the loss of data and essential information of organizations in e-Government. In this paper, we introduce a holistic approach to assessing information security risks based on both qualitative and quantitative methods for the Vietnamese e-Government. Our model of security risk management is built according to both international standards (ISO 27005-2018, NIST SP800-30r1, SP800-39, SP800-53r4) and Vietnamese standard (TCVN). For the quantitative risk method, we use both CVSS and OWASP scoring standards to quantify information system risks. Besides, the information security risks of the system can also be determined through vulnerability scanners. We also implemented the proposed model in a Web application, called SoC.UET. The experiments we conducted with UET.SoC allowed proving the ability to manage the information security risks holistically for a Ministry or a Province in the Vietnamese e-Governmen %0 Journal Article %@ 1813-9663 %A Le, Vinh %D 2019 %F SisLab:3525 %I Vietnam Academy of Science & Technology %J Journal of Computer Science and Cybernetics %N 14 %P 105-118 %T A computational framework to analyze human genomes %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3525/ %V 35 %X The advent of genomic technologies have led to the genomic era. The human genome consists of more than 3 billions of nucleotides and the large scale human genome projects resulted in a huge amount of genomic data. Analyzing human genomes is a challenging computational procedure including a number of key steps from aligning short reads, calling variants, and annotating variants. This paper describes a framework to analyze human genomes from the raw data to final results that can be used for diagnosing diseases in clinical tests. We also analyzed 101 Vietnamese human genomes to propose a list of nucleotide sites that might result in false positive results when calling variants from single or few genomes. %0 Journal Article %@ 1098-1004 %A Le, Vinh %A Tran, Kien %A Bui, Hoa %A Le, Huong %A Nguyen, Canh %A Do, Duong %A Ly, Ha %A Pham, Ly %A Dao, Lan %A Nguyen, Liem %D 2019 %F SisLab:3733 %I Wiley %J Human Mutation %N 10 %P 1664-1675 %T A Vietnamese human genetic variation database %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3733/ %V 40 %X Large scale human genome projects have created tremendous human genome databases for some well‐studied populations. Vietnam has about 95 million people (the 14th largest country by population in the world) of which more than 86% are Kinh people. To date, genetic studies for Vietnamese people mostly rely on genetic information from other populations. Building a Vietnamese human genetic variation database is a must for properly interpreting Vietnamese genetic variants. To this end, we sequenced 105 whole genomes and 200 whole exomes of 305 unrelated Kinh Vietnamese (KHV) people. We also included 101 other previously published KHV genomes to build a Vietnamese human genetic variation database of 406 KHV people. The KHV database contains 24.81 million variants (22.47 million single nucleotide polymorphisms (SNPs) and 2.34 million indels) of which 0.71 million variants are novel. It includes more than 99.3% of variants with a frequency of >1% in the KHV population. Noticeably, the KHV database revealed 107 variants reported in the human genome mutation database as pathological mutations with a frequency above 1% in the KHV population. The KHV database (available at https://genomes.vn) would be beneficial for genetic studies and medical applications not only for the Vietnamese population but also for other closely related populations. %0 Conference Paper %A Le, Vu Ha %B IEEE International Conference on Image Processing %C Singapore %D 2004 %F SisLab:1851 %P 1955-1958 %T A structure-from-motion method for 3-D reconstruction of moving objects from multiple-view image sequences %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1851/ %V 3 %0 Book %A Le, Vu Ha %A Dang, Minh Quy %C Hanoi %D 2014 %E Le, Vu Ha %F SisLab:2708 %I VNU Publisher %P 150 %T A Guidebook to Principles of Emerging Information and Communication Technologies %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2708/ %0 Conference Paper %A Le, Vu Ha %A Gouiffes, Michele %A Parrain, Fabien %A Bosseboeuf, Alain %A Zavidovique, Bertrand %B SPIE Optical Engineering + Applications %C San Diego, CA, USA %D 2007 %F SisLab:2364 %P 66962D-66962D %T Image blur analysis for the subpixel-level measurement of in-plane vibration parameters of MEMS resonators %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2364/ %V 6696 %0 Journal Article %@ 0922-5773 %A Le, Vu Ha %A Seetharaman, Guna %D 2006 %F SisLab:1845 %J Journal of VLSI signal processing systems for signal, image and video technology %N 1 %P 79-89 %T A Super-Resolution Imaging Method Based on Dense Subpixel-Accurate Motion Fields %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1845/ %V 42 %0 Conference Paper %A Le, Vu Ha %A Seetharaman, Guna %B International Workshop on Digital and Computational Video %C Clearwater Beach, FL, USA %D 2002 %F SisLab:1854 %P 35-42 %T A superresolution imaging method based on dense subpixel-accurate motion fields %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1854/ %0 Conference Paper %A Le, Vu Ha %A Seetharaman, Guna %A Zavidovique, Bertrand %B IEEE International Workshop on Computer Architectures for Machine Perception %C New Orleans, LA, USA %D 2003 %F SisLab:1852 %P 221-230 %T A Multiscale Technique for Optical Flow Computation Using Piecewise Affine Approximation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1852/ %0 Journal Article %A Le, Dinh Thanh %D 2013 %F SisLab:782 %J Journal of Science of HNUE %T More Efficient Path Optimization for Greedy Geographic Routing in Wireless Sensor Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/782/ %V 58 %0 Journal Article %A Le Dinh, Thanh %A Nguyen, Dai Tho %D 2016 %F SisLab:2007 %J Research, Development and Application on Information & Communication Technology %N 13 %T Indoor Fire Detection using Wireless Sensor Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2007/ %V 3 %X In this paper, we propose an algorithm for early detection of indoor fire based on temperature and light data collected by wireless sensor networks. Our goal is to develop an affordable, acceptable, detection method based on wireless sensor networks with cheap and highly available sensors. Experimental results show that our proposed algorithm is able to detect fire quickly and accurately. %0 Book %A Le Dinh, Thanh %A Nguyen, Viet Anh %C Hanoi, Vietnam %D 2018 %F SisLab:3079 %I VNU Publishing House %P 220 %T Phát triển ứng dụng web %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3079/ %0 Conference Paper %A Le Trung, Thanh %A Karim, Abed-Meraim %A Nguyen Linh, Trung %A Remy, Boyer %B 28th European Signal Processing Conference (EUSIPCO) %C Amsterdam, The Netherlands %D 2021 %F SisLab:4003 %T Adaptive Algorithms for Tracking Tensor-Train Decomposition of Streaming Tensors %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4003/ %X Tensor-train (TT) decomposition has been an efficient tool to find low order approximation of large-scale, high-order tensors. Existing TT decomposition algorithms are either of high computational complexity or operating in batch-mode, hence quite inefficient for (near) real-time processing. In this paper, we propose a novel adaptive algorithm for TT decomposition of streaming tensors whose slices are serially acquired over time. By leveraging the alternating minimization framework, our estimator minimizes an exponentially weighted least-squares cost function in an efficient way. The proposed method can yield an estimation accuracy very close to the error bound. Numerical experiments show that the proposed algorithm is capable of adaptive TT decomposition with a competitive performance evaluation on both synthetic and real data. %0 Journal Article %@ 1225-6463 %A Lee, Yang Koo %A Vu, Thi Hong Nhan %A Le, Thanh Ha %D 2015 %F SisLab:1477 %I ETRI %J ETRI Journal %N 2 %P 222-232 %T Dual-phase approach to Improve Prediction of Heart Disease in Mobile Environment %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1477/ %V 37 %0 Journal Article %@ 2169-3536 %A Li, Zhuorong %A Yu, Hongchuan %A Kieu, Hai Dang %A Vuong, Tung Long %A Zhang, Jian Jun %D 2020 %F SisLab:4155 %I IEEE %J IEEE Access %T PCA-Based Robust Motion Data Recovery %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4155/ %X Human motion tracking is a prevalent technique in many fields. A common difficulty encountered in motion tracking is the corrupted data is caused by detachment of markers in 3D motion data or occlusion in 2D tracking data. Most methods for missing markers problem may quickly become ineffective when gaps exist in the trajectories of multiple markers for an extended duration. In this paper, we propose the principal component eigenspace based gap filling methods that leverage a training sample set for estimation. The proposed method is especially beneficial in the scenario of motion data with less predictable or repeated movement patterns, and that of even missing entire frames within an interval of a sequence. To highlight algorithm robustness, we perform algorithms on twenty test samples for comparison. The experimental results show that our methods are numerical stable and fast to work. %0 Conference Paper %A Lim, Kwang Hyeon %A Lee, Yang Koo %A Vu, Thi Hong Nhan %A Bae, Kyong Ho %A Heo, Min %B 4th International Conference FITAT/DBMI 2011 %C Korea %D 2011 %F SisLab:2133 %T Query Processing Algorithm for Spatiotemporal Sensor Data Stream %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2133/ %0 Conference Paper %A Lu, Dang Nhac %A Ngo, Thi Thu Trang %A Nguyen, Duc Nhan %A Nguyen, Thi Hau %A Nguyen, Ha Nam %B 2017 International Conference on Information Science and Applications %D 2017 %F SisLab:2584 %T A Novel Mobile Online Vehicle Status Awareness Method Using Smartphone Sensors %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2584/ %X In this paper, we proposed an efficient method with flexible framework for vehicle status awareness using smartphone sensors, so called Mobile Online Vehicle Status Awareness System (MOVSAS). The system deployed while users to put their smartphones in any position and at any direction. In our proposed framework, principal component analysis (PCA) algorithm is used to selected suitable features from set of combining features on time-base, power-based and frequency-based domain, which extracted from accelerometer sensor data. The classification model using Random Forest (RF), Naïve Bayes (NB), K-Nearest Neighbor (KNN), and Support Vector Machine (SVM) algorithms to deploy for awareness issues of vehicle status. The refining model is proposed using Artificial Neural Network (ANN) algorithm aim to improved accuracy prediction vehicle status results before. Training data sets, which are collected and the dynamic feedback also helping improved accuracy of system. A number of experiments are shown that the high accuracy of MOVSAS with vehicle kinds as bicycle, motorbike and car. %0 Journal Article %@ 1424-8220 %A Lu, Dang Nhac %A Nguyen, Duc Nhan %A Nguyen, Thi Hau %A Nguyen, Ha Nam %D 2018 %F SisLab:2929 %I MDPI %J Sensors %N 1036 %T Vehicle Mode and Driving Activity Detection Based on Analyzing Sensor Data of Smartphones %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2929/ %V 18 %X In this paper, we present a flexible combined system, namely the Vehicle mode-driving Activity Detection System (VADS), that is capable of detecting either the current vehicle mode or the current driving activity of travelers. Our proposed system is designed to be lightweight in computation and very fast in response to the changes of travelers’ vehicle modes or driving events. The vehicle mode detection module is responsible for recognizing both motorized vehicles, such as cars, buses, and motorbikes, and non-motorized ones, for instance, walking, and bikes. It relies only on accelerometer data in order to minimize the energy consumption of smartphones. By contrast, the driving activity detection module uses the data collected from the accelerometer, gyroscope, and magnetometer of a smartphone to detect various driving activities, i.e., stopping, going straight, turning left, and turning right. Furthermore, we propose a method to compute the optimized data window size and the optimized overlapping ratio for each vehicle mode and each driving event from the training datasets. The experimental results show that this strategy significantly increases the overall prediction accuracy. Additionally, numerous experiments are carried out to compare the impact of different feature sets (time domain features, frequency domain features, Hjorth features) as well as the impact of various classification algorithms (Random Forest, Naïve Bayes, Decision tree J48, K Nearest Neighbor, Support Vector Machine) contributing to the prediction accuracy. Our system achieves an average accuracy of 98.33% in detecting the vehicle modes and an average accuracy of 98.95% in recognizing the driving events of motorcyclists when using the Random Forest classifier and a feature set containing time domain features, frequency domain features, and Hjorth features. Moreover, on a public dataset of HTC company in New Taipei, Taiwan, our framework obtains the overall accuracy of 97.33% that is considerably higher than that of the state-of the art. %0 Conference Paper %A Lu, Dang Nhac %A Nguyen, Thi Hau %A Ngo, Thi Thu Trang %A Nguyen, Duc Nhan %A Nguyen, Ha Nam %B 2017 International Conference on Information Networking (ICOIN) %D 2017 %F SisLab:2487 %P 584-589 %T A novel traffic routing method using hybrid Ant Colony System based on genetic algorithm %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2487/ %0 Journal Article %@ 2194-5357 %A Lu, Dang Nhac %A Nguyen, Thu Trang %A Ngo, Thi Thu Trang %A Nguyen, Thi Hau %A Nguyen, Ha Nam %D 2016 %F SisLab:1897 %J Advances in Information and Communication Technology %P 357-366 %T Mobile Online Activity Recognition System Based on Smartphone Sensors %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1897/ %V 538 %X In this paper, we propose an efficient and flexible framework for activity recognition based on smartphone sensors. We develop a mobile application that integrates data collection, training and recognition, feedback monitoring. This system allows user smartphones are randomly placed in any position and at any direction. In the proposed framework, a set of power based and frequency-based features is extracted from sensor data. Then, we deploy Random Forest, Naïve Bayes, K-Nearest Neighbor (KNN), Support Vector Ma-chine (SVM) classification algorithms for recognizing a set of user activities. Our framework dynamically takes into account real-time user feedbacks to increase the prediction accuracy. Our framework will be able to apply for intelligent mo-bile applications. A number of experiments were carried out to show the high ac-curacy of the proposed framework for detecting user activities when walking or driving a motorbike. %0 Journal Article %A Lu, Dang Nhac %A Nguyen, Thu Trang %A Nguyen, Thi Hau %A Nguyen, Ha Nam %A Choi, Gyoo Seok %D 2015 %F SisLab:1729 %J International Journal of Internet, Broadcasting and Communication %N 1 %P 20-22 %T Detecting smartphone user habits using sequential pattern analysis %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1729/ %V 7 %0 Conference Paper %A Lu, Dang Nhac %A Tran, Thuy Binh %A Nguyen, Duc Nhan %A Nguyen, Thi Hau %A Nguyen, Ha Nam %B 3rd EAI International Conference on Nature of Computation and Communication 2017 %D 2017 %F SisLab:2696 %T Abnormal Behavior Detection Based on Smartphone Sensors %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2696/ %X There are a lot of applications were developed to take advance of smartphone sensors for utilizing the personal services such as health-care, walk-counting, routing etc. Users behavior analysis is attracted a lot of researches in-terested with various approaches. We proposed a novel framework to detect the abnormal driving behavior using smartphone sensors. It named Abnormal Be-havior Detection System (ABDS). The system keep track the driver activities during he’s trip based on smartphone sensors. The Practice Swarm Optimiza-tion (PSO) algorithm is used to automatically select suitable features extracted from sensors data. The oriented accelerometer is used to detect activity. The abnormal behavior is collected and labeled then detection by Artificial Neural Network (ANN). The implementation shown the promising results in case of seven activities (stop, moving, acceleration, deceleration, turn left, turn right and U-turn) with 86.71% accuracy. %0 Conference Paper %A Luong, Thai Le %A Cao, Minh Son %A Le, Duc Thang %A Phan, Xuan Hieu %B The 9th International Conference on Knowledge and Systems Engineering (KSE 2017) %C Hue city, Vietnam %D 2017 %F SisLab:2594 %P 215-220 %T Intent Extraction from Social Media Texts Using Sequential Segmentation and Deep Learning Models %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2594/ %X Nowadays, users are much more willing to share their daily activities, their thoughts or feelings, and even their intentions (e.g., buy an apartment, rent a car, travel to somewhere, etc.) on online social media channels. Understanding intents of online users, therefore, has become a crucial need in many different business areas like production, finance/banking, real estate, tourism, e-commerce, and online marketing. In this paper, we will present our solutions to extract intent information from online social media texts. This can be seen as an information extraction or sequential segmentation task. In order to perform this task, we have built our machine learning models based on conditional random fields (CRFs), an advanced statistical graphical model for sequence data, and bidirectional long short–term memory (Bi–LSTM), a well-known deep learning model. To evaluate our methods, we have defined two intent extraction tasks in two domains: real estate and cosmetics & beauty. For each task, we have defined the tag set as well as prepared the labeled data set that consists of Vietnamese social media text posts with the annotation of intent words/phrases. Experimental results showed that the proposed methods can effectively extract intent information from online texts with significantly high accuracy. %0 Journal Article %@ 1405-5546 %A Luong, Thai Le %A Tran, Nhu Thuat %A Dang, Tien Son %A Tran, Quoc Long %A Phan, Xuan Hieu %D 2020 %F SisLab:3677 %J Computacion y Sistemas %N 1 %T Domain-independent intent extraction from online texts %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3677/ %V 24 %X Identifying user’s intents from texts on online channels has a wide range of applications from entrepreneurship, banking to e-commerce. However, intent identification is not a simple task due to the intent and its attributes are various and strongly depend on the domain of data. If the number of intent domains increases, the number of intent’s attributes will get bigger. As a result, the complexity of intent extraction task grows up significantly. Additionally, when a new domain comes, it involves considerable physical efforts to define specific labels for intent and attributes for that domain. Hence, it would be much better to come up with a new method for extracting user’s intents which is not dependent on a specific domain. In our research, we study the problem of domain-independent intent identification from posts and comments crawled from social networks and discussion forums. We present ten general labels, i.e. labels do not depend on a specific domain, and utilize them when extracting intent and its related information. We also propose a map between general labels and domain-specific labels. We extensively conduct experiments to explore the efficiency of using general labels compared to specific labels in extracting user’s intents when the number of intent domains increases. Our study is conducted on a medium-sized dataset from three selected domains: Tourism, Real Estate and Transportation. In term of accuracy, when the number of domains grows, our proposal achieves significantly better results than domain-specific method in identifying user's intent. %0 Conference Paper %A Luong, Thai Le %A Tran, Nhu Thuat %A Phan, Xuan Hieu %B The 19th International Symposium on Communications and Information Technologies (ISCIT 2019) %C Ho Chi Minh City, Vietnam %D 2019 %F SisLab:3642 %P 58-63 %T Improving Intent Extraction Using Ensemble Neural Network %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3642/ %X User intent extraction from social media texts is aimed at identifying user intent keyword and its related information. This topic has attracted a lot of researches since its various applications in online marketing, e-commerce and business services. One of such studies is to model this problem as a sequence labeling task and apply state-of-the-art sequential tagging models such as BiLSTM [12] and BiLSTM-CRFs [12]. In this paper, we take a further step to enhance intent extraction results based on tri-training [23] and ensemble learning [2]. Specifically, we simultaneously use three BiLSTM-CRFs models, each of them is different from others by the type of word embeddings, and apply majority voting scheme over their predicted labels when decoding final labels. Extensive experiments on data from three domains Real Estate, Tourism and Transportation show that our proposed methods enjoy a better performance compared to single model based approach. %0 Conference Paper %A Luong, Thai Le %A Tran, Thi Hanh %A Truong, Quoc Tuan %A Truong, Thi Minh Ngoc %A Phi, Thi Thu %A Phan, Xuan Hieu %B The 8th Asian Conference on Intelligent Information and Database Systems (ACIIDS) %C Da Nang, Vietnam %D 2016 %F SisLab:1883 %P 14-16 %T Learning to Filter User Explicit Intents in Online Vietnamese Social Media Texts %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1883/ %V 9622 %X Today, Internet users are much more willing to express themselves on online social media channels. They commonly share their daily activities, their thoughts or feelings, and even their intention (e.g., buy a camera, rent an apartment, borrow a loan, etc.) about what they plan to do on blogs, forums, and especially online social networks. Understanding intents of online users, therefore, has become a crucial need for many enterprises operating in different business areas like production, banking, retail, e–commerce, and online advertising. In this paper, we will present a machine learning approach to analyze users’ posts and comments on online social media to filter posts or comments containing user plans or intents. Fully understanding user intent in social media texts is a complicated process including three major stages: user intent filtering, intent domain identification, and intent parsing and extraction. In the scope of this study, we will propose a solution to the first one, that is, building a binary classification model to determine whether a post or comment carries an intent or not. We carefully conducted an empirical evaluation for our model on a medium-sized collection of posts in Vietnamese and achieved promising results with an average accuracy of more than 90%. %0 Conference Paper %A Luong, Thai Le %A Trang, Thi Hanh %A Truong, Quoc Tuan %A Truong, Thi Minh Ngoc %A Phi, Thi Thu %A Phan, Xuan Hieu %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1527 %T Learning to Filter User Explicit Intents in Online Vietnamese Social Media Texts %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1527/ %X Today, Internet users are much more willing to express them- selves on online social media channels. They commonly share their daily activities, their thoughts or feelings, and even their intention (e.g., buy a camera, rent an apartment, borrow a loan, etc.) about what they plan to do on blogs, forums, and especially online social networks. Understand- ing intents of online users, therefore, has become a crucial need for many enterprises operating in different business areas like production, banking, retail, e–commerce, and online advertising. In this paper, we will present a machine learning approach to analyze users’ posts and comments on online social media to filter posts or comments containing user plans or intents. Fully understanding user intent in social media texts is a compli- cated process including three major stages: user intent filtering, intent domain identification, and intent parsing and extraction. In the scope of this study, we will propose a solution to the first one, that is, building a binary classification model to determine whether a post or comment carries an intent or not. We carefully conducted an empirical evaluation for our model on a medium–sized collection of posts in Vietnamese and achieved promising results with an average accuracy of more than 90%. Keywords: Intention mining, user intent identification, social media text understanding, content filtering, text classification %0 Conference Paper %A Luong, Thai Le %A Truong, Quoc Tuan %A Dang, Hai Trieu %A Phan, Xuan Hieu %B The 7th Symposium on Information and Communication Technology (SoICT) %C Ho Chi Minh City, Viet Nam %D 2016 %F SisLab:2292 %P 52-57 %T Domain identification for intention posts on online social media %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2292/ %0 Conference Paper %A Luong, Thanh Nhan %A To, Van Khanh %A Truong, Ninh Thuan %B 9th Asian Conference on Intelligent Information and Database Systems %C Kanazawa, Japan %D 2017 %F SisLab:2541 %I Springer %P 489-498 %T Checking Compliance of Program with Secure UML Model %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2541/ %0 Conference Paper %A Luong, Thanh Nhan %A Truong, Ninh Thuan %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1528 %T Phương pháp kiểm chứng các tính chất an ninh của phần mềm %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1528/ %X Trong giai đoạn kỹ thuật công nghệ phát triển như hiện nay, an ninh là yêu cầu quan trọng đối với hệ thống chăm sóc sức khỏe vì nhu cầu cần bảo vệ các thông tin riêng tư của bệnh nhân. Việc tích hợp các chính sách an ninh vào hệ thống phần mềm này là rất cần thiết để giảm thiểu các vi phạm an ninh. Kiểm soát truy cập dựa trên vai trò là phương pháp hiệu quả để thực hiện các tính chất an ninh cơ bản của một phần mềm an toàn. Nghiên cứu trình bày một phương pháp để đặc tả các yêu cầu an ninh trong hệ thống khám chữa bệnh. Bên cạnh việc đề xuất một phương pháp chứng minh sự đáp ứng của các yêu cầu an ninh với một hệ thống phần mềm này, một công cụ kiểm chứng cũng được giới thiệu. %0 Conference Paper %A Luong, Thanh Nhan %A Vo, Dinh Hieu %A To, Van Khanh %A Truong, Ninh Thuan %B ICCASA 2018 - 7th EAI International Conference on Context-Aware Systems and Applications %C Vietnam %D 2018 %F SisLab:3308 %T On The Compliance Of Access Control Policies In Web Applications %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3308/ %0 Conference Paper %A Luong, Thanh Nhan %A Vo, Dinh Hieu %A Truong, Ninh Thuan %B The NAFOSTED Conference on Information and Computer Science (NICS) %D 2019 %F SisLab:3637 %T An Approach to Analyze Software Security Requirements in ABAC Mode %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3637/ %X Security has been a crucial aspect of most applications especially, critical-safety softwares. In fact, losing or leaking of sensitive data can lead to huge losses for organizations so software developers must always find ways to ensure the security properties for their products. In practice, attribute-based access control (ABAC) has been an effective, flexible and popular method to mitigate the risks of unauthorized accesses to resources in large and complex systems. Therefore, we introduce an approach for checking ABAC rules from source code of an application software against to its requirement specification. Our work includes of a formal definition about ABAC policy, method to implement ABAC into application as well as analyze access rules from the source code, and a algorithm to check analyzed ABAC rules against to its specification. The proposed approach can help programmers to detect the inconsistency between specification and implementation. We also illustrate our approach with an example in a medical information management system. %0 Conference Paper %A Luong, Thi Thuy %A Nguyen, Viet Thang %A Pham, Minh Trien %B 2018 REV-ECIT: National Conference on Electronics, Communications and Information Technology %C Hanoi, Vietnam %D 2018 %F SisLab:3247 %P 38-42 %T Ảnh Hưởng Của Giao Tiếp Đến Đánh Giá Lỗi Robot Trong Tương Tác Người – Robot %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3247/ %X Chúng tôi tiến hành một nghiên cứu hành vi con người mà chúng tôi đã cố tình lập trình hành vi lỗi cho robot. Mục đích của nghiên cứu để khám phá sự khác nhau trong đánh giá của người dùng về giá trị của giao tiếp trong tương tác với robot. Những người tham gia nghiên cứu được hướng dẫn bởi một robot giả định lập trình trên máy vi tính để thực hiện nhiệm vụ nấu một món ăn trong 3 kịch bản có sẵn. Kịch bản thứ nhất hướng dẫn người tham gia nấu ăn trong điều kiện không có lỗi nhưng robot chỉ hướng dẫn mà không giao tiếp. Kịch bản thứ hai robot hướng dẫn người tham gia trong điều kiện có lỗi. Kịch bản thứ ba robot hướng dẫn người tham gia trong điều kiện có lỗi nhưng sử dụng các yếu tố giao tiếp trong quá trình hướng dẫn. Sau tương tác, chúng tôi yêu cầu những người tham gia đánh giá độ khó của nhiệm vụ và mức độ hài lòng của họ. Kết quả cho thấy mặc dù có lỗi nhưng một robot biết giao tiếp được đánh giá cao hơn so với robot bình thường. %0 Conference Paper %A Luong, Viet Nguyen %A Ho, Van Canh %A Trinh, Nhat Tien %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1552 %T Nghiên cứu và đề xuất một phương pháp bảo vệ bản quyền tài liệu số %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1552/ %X Nghiên cứu đề xuất phương pháp được phát triển trong đó một hình ảnh thủy vân xuất hiện trong một bản in và sau đó có thể được trích xuất bằng cách sử dụng máy quét để quét bản in và một phần mềm để khai thác thủy vân. Trong trường hợp này, phương pháp nhúng phải được thiết kế cẩn thận bởi vì cả hai quá trình in ấn và chụp xuất hiện nguyên nhân biến dạng mà làm cho khai thác thủy vân là một thách thức. Trọng tâm của nghiên cứu là kỹ thuật thủy vân mù, thông tin được nhúng là nhiều bít, đánh giá sự bền vững của các thuật toán được thử nghiệm trong một môi trường văn phòng, sử dụng thiết bị văn phòng. %0 Journal Article %@ 0866-8612 %A Luong, Viet Nguyen %A Trinh, Nhat Tien %A Ho, Van Canh %D 2016 %F SisLab:2106 %I VNU Journal of Science %J VNU Journal of Computer Science and Communication Engineering %N 2 %P 49-62 %T A Watermark Algorithm Against De-Synchronization Attacks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2106/ %V 32 %X In this paper, a robust method to the ability of the watermark to resist against attacks is proposed for hiding information into images. The proposed method is blind because the original image is not required at the decoder to recover the embedded data. The robustness of the watermarking scheme is inspired by using a PJND (Pyramidal Just Noticeable Difference) model and the message is inserted into these DoG (Difference of Gaussians) [1, 2]. Our proposal takes into account three main characteristics of Human Visual System, namely: contrast sensitivity, luminance adaptation and contrast marking. Therefore, it not only provides an invisible and robust watermarking but also optimizes watermarking capacity. The performance of the proposed technique is evaluated by a series of experiments with different input images. In terms of transparency, besides using the subjective experiments, eight objective metrics are calculated in comparison with other methods such as PSNR, MSSIM, SVDm, etc. Our approach always presents the outperform values. In terms of robustness, many kinds of attacks from global transformation (rotation, scaling, etc) to local transformation (stirmark, checkmark benchmarks, de-synchronization attacks) are implemented. Many image processing tools are applied to simulate the attacks such as Print-Screen, Using Photo editing software, Camcorder, Print-Scan, etc. The experimental results show an outstanding robustness in resisting these attacks. %0 Journal Article %A Luu, Manh Ha %A Walsum, Theo van %A Mai, Hong Son %A Franklin, Daniel %A Nguyen, Thi Thu Thao %A Le, Thi My %A Moelker, Adraan %A Le, Van Khang %A Vu, Dang Luu %A Le, Ngoc Ha %A Tran, Quoc Long %A Chu, Duc Trinh %A Nguyen, Linh Trung %D 2022 %F SisLab:4765 %J Medical Image Analysis %T Automatic scan range for dose-reduced multiphase CT imaging of the liver utilizing CNNs and Gaussian models %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4765/ %X Multiphase CT scanning of the liver is performed for several clinical applications; however, radiation exposure from CT scanning poses a nontrivial cancer risk to the patients. The radiation dose may be reduced by determining the scan range of the subsequent scans by the location of the target of interest in the first scan phase. The purpose of this study is to present and assess an automatic method for determining the scan range for multiphase CT scans. Our strategy is to first apply a CNN-based method for detecting the liver in 2D slices, and to use a liver range search algorithm for detecting the liver range in the scout volume. The target liver scan range for subsequent scans can be obtained by adding safety margins achieved from Gaussian liver motion models to the scan range determined from the scout. Experiments were performed on 657 multiphase CT volumes obtained from multiple hospitals. The experiment shows that the proposed liver detection method can detect the liver in 223 out of a total of 224 3D volumes on average within one second, with mean intersection of union, wall distance and centroid distance of 85.5%, 5.7 mm and 9.7 mm, respectively. In addition, the performance of the proposed liver detection method is comparable to the best of the state-of-the-art 3D liver detectors in the liver detection accuracy while it requires less processing time. Furthermore, we apply the liver scan range generation method on the liver CT images acquired from radiofrequency ablation and Y-90 transarterial radioembolization (selective internal radiation therapy) interventions of 46 patients from two hospitals. The result shows that the automatic scan range generation can significantly reduce the effective radiation dose by an average of 14.5% (2.56 mSv) compared to manual performance by the radiographer from Y-90 transarterial radioembolization, while no statistically significant difference in performance was found with the CT images from intra RFA intervention ( = 0.81). Finally, three radiologists assess both the original and the range-reduced images for evaluating the effect of the range reduction method on their clinical decisions. We conclude that the automatic liver scan range generation method is able to reduce excess radiation compared to the manual performance with a high accuracy and without penalizing the clinical decision. %0 Conference Paper %A Luu, Quang Thang %A Nguyen, Thi Nhat Thanh %A Nguyen, Thanh Thuy %A Bui, Quang Hung %B The 9th International Conference on Knowledge and Systems Engineering (KSE) %C Hue, Vietnam %D 2017 %F SisLab:2775 %T TMACT: A Thematic Map Automatically Creating Tool For Maintaining WebGIS Systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2775/ %0 Conference Paper %A Luu, Quang Thang %A Phan, Anh %A Luu, Viet Hung %A Nguyen, Thi Nhat Thanh %A Bui, Quang Hung %B Hội thảo khoa học Ứng dụng GIS toàn quốc 2017 %D 2017 %F SisLab:2773 %P 883-892 %T Sử dụng công nghệ Hadoop và Spark để lưu trữ và xử lý phân tán ảnh vệ tinh VNREDSat-1 %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2773/ %0 Conference Paper %A Luu, Quang Thang %A Phan, Anh %A Nguyen, Thi Nhat Thanh %A Bui, Quang Hung %B Hội thảo khoa học Ứng dụng GIS toàn quốc 2017 %C Quy Nhơn, Viet Nam %D 2017 %F SisLab:2774 %P 874-882 %T SEAP: nền tảng mới cho việc khai thác và phân tích dữ liệu không gian lớn %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2774/ %0 Conference Paper %A Luu, Viet Hung %A Dao, Minh Son %A Nguyen, Thi Nhat Thanh %A Perry, Stuart %A Zettsu, Koji %B 6th NAFOSTED Conference on Information and Computer Science %C Hanoi %D 2019 %F SisLab:3582 %T Semi-supervised Convolutional Neural Networks for Flood Mapping using Multi-modal Remote Sensing Data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3582/ %X When floods hit populated areas, quick detection of flooded areas is crucial for initial response by local government, residents, and volunteers. Space-borne polarimetric synthetic aperture radar (PolSAR) is an authoritative data sources for flood mapping since it can be acquired immediately after a disaster even at night time or cloudy weather. Conventionally, a lot of domain specific heuristic knowledge has been applied for PolSAR flood mapping, but their performance still suffers from confusing pixels caused by irregular reflections of radar waves. Optical images are another data source that can be used to detect flooded areas due to their high spectral correlation with the open water surface. However, they are often affected by day, night, or severe weather conditions (i.e., cloud). This paper presents a convolution neural network (CNN) based multimodal approach utilizing the advantages of both PolSAR and optical images for flood mapping. First, reference training data is retrieved from optical images by manual annotation. Since clouds may appear in the optical image, only areas with a clear view of flooded or non-flooded are annotated. Then, a semisupervised polarimetric features-aided CNN is utilized for flood mapping using PolSAR data. The proposed model not only can handle the issue of learning with incomplete ground truth but also can leverage a large portion of unlabelled pixels for learning. Moreover, our model takes the advantages of expert knowledge on scattering interpretation to incorporate polarimetric-features as the input. Experiments results are given for the flood event that occurred in Sendai, Japan, on 12th March 2011. The experiments show that our framework can map flooded area with high accuracy (F1=96.12) and outperform conventional flood mapping methods. %0 Journal Article %@ 2150-704X %A Luu, Viet Hung %A Dinh, Van Kiet %A Luong, Nguyen Hoang Hoa %A Bui, Quang Hung %A Nguyen, Thi Nhat Thanh %D 2019 %F SisLab:3449 %I Taylor & Francis %J Remote Sensing Letters %N 9 %P 826-834 %T Improving the Bag-of-Words model with Spatial Pyramid matching using data augmentation for fine-grained arbitrary-oriented ship classification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3449/ %V 10 %X In this letter, we investigate fine-grained classification of arbitrary-oriented ships in very high resolution optical imagery using Bag of Word model with Spatial Pyramid (SP-BoW). Given that based on ‘spatial pyramid’ of the histogram of local features, the final feature vectors not only count the multiplicity of ‘words’ but also represent their spatial topology. We attempt to improve the performance of this model by introducing augmented data for training phase. Our aim is to make the dataset big enough to be able to capture holistic variation of ship orientation. Three data augmentation operations are used including random rotate by an angle of modulo 90°, random flip-left-right, and random flip-top-bottom. Through this procedure, our trained SP-BoW model is able to get better generalization. The proposed approach is validated on the High-Resolution Ship Collections 2016 (HRSC2016) ship dataset. The results indicate that training on augmented data can significantly improve the performance of SP-BoW. Beside, compared to other state-of-the-art convolutional neural network-based approaches, the approach proposed in this research has yielded competitive results and could make it a good baseline for evaluating more sophisticated CNN architecture in the future. %0 Conference Paper %A Luu, Viet Hung %A Man, Duc Chuc %A Luong, Chinh Ke %A Bui, Quang Hung %A Nguyen, Thi Nhat Thanh %B The 2nd International Joint Conference (IJCC) %C Hanoi, Vietnam %D 2016 %F SisLab:2105 %P 130-135 %T Air pollution mapping from high spatial resolution satellite images: a case study in Hanoi %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2105/ %0 Journal Article %A Luu, Viet Hung %A Pham, Van Manh %A Man, Duc Chuc %A Bui, Quang Hung %A Nguyen, Thi Nhat Thanh %D 2016 %F SisLab:2097 %I The International Promotion Agency of Culture Technology %J International Journal of Advanced Culture Technology (IJACT) %N 2 %P 1-6 %T Comparison of Various Image Fusion Methods for Impervious Surface Classification from VNREDSat-1 %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2097/ %V 4 %X Impervious surfaces are important indicators for urban development monitoring. Accurate mapping of urban impervious surfaces with observational satellites, such as VNREDSat-1, remains challenging due to the spectral diversity not captured by an individual PAN image. In this article, five multi-resolution image fusion techniques were compared for the task of classifting urban impervious surfaces. The result shows that for VNREDSat-1 dataset, UNB and Wavelet tranformation methods are the best techniques in reserving spatial and spectral information of original MS image, respectively. However, the UNB technique gives the best results when it comes to impervious surface classification, especially in the case of shadow areas included in non-impervious surface group. %0 Conference Paper %A Ma, Thi Chau %A Bui, The Duy %A Dang, Trung Kien %A VNU_UET, %B 2012 Fourth International Conference on Knowledge and Systems Engineering (KSE) %C Danang, Vietnam %D 2012 %F SisLab:104 %K Accuracy;Biological tissues;Cameras;Face;Feature extraction;Image reconstruction;Shape;computer graphics;image reconstruction;3D final model;image based 3D skull feature reconstruction;shift error analysis;shift errors;skull images;skull-based 3D facial reconstruction;3D facial reconstruction;accuracy;feature detection; %P 4 -10 %T Shift Error Analysis in Image Based 3D Skull Feature Reconstruction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/104/ %X 3D skull is crucial in skull-based 3D facial reconstruction. In 3D reconstruction, especially in skull-based 3D facial reconstruction, features usually play an important role. Because, the accuracy in feature detection strongly affects the accuracy of the 3D final model. In this paper, we concentrate on accuracy of 3D reconstructed skull, one important part in skull-based 3D facial reconstruction. We discuss a cause of errors called shift errors when taking sequence of skull images. In addition, we analysis the effect of shift error in 3D reconstruction and propose solution to limit the effect. %0 Conference Paper %A Ma, Thi Chau %A Chetcuti-Sperandiox, Nathalie %A Lagruex, Sylvain %A Nguyen, Thanh Thuy %A Bui, The Duy %A VNU-UET, %B International Conference on Digital Arts, Media and Technology (ICDAMT) %C Chiengmai, Thailand %D 2017 %F SisLab:2646 %T Towards an Ontology for Vietnamese Water Puppetry %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2646/ %X In this paper we propose the first steps for building a Vietnamese water puppetry ontology. The creation of the puppetry ontology would allow not only a more informed and productive professional training, but also the capability of preserving and promoting water puppetry. Gathering water puppetry expert knowledge requires a deep study of puppetry history, background knowledge and performance. Hence, we analyze stories, background knowledge and performance of the water puppetry in the context of the Vietnamese general culture and legends. Especially, we emphasize on the issue of inconsistencies - a key challenge in ontology building. We also present a specific case study using DL-Lite for representation, reasoning and querying in presence of inconsistencies. %0 Journal Article %@ 0866-8612 %A Ma, Thi Chau %A Hoang, Minh Duong %D 2016 %F SisLab:2028 %I VNU Journal of Science %J VNU Journal of Computer Science and Communication Engineering %N 2 %P 23-30 %T Kinect based character navigation in VR Game %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2028/ %V 32 %0 Conference Paper %A Ma, Thi Chau %A Nguyen, Dinh Tu %A Dinh, Quang Huy %A Bui, The Duy %A Nguyen Dinh Tu, %A Dinh Quang Huy, %A Bui The Duy, %B KSE: the 2011 International Conference on Knowledge and Systems Engineering %C Vietnam %D 2011 %F SisLab:2652 %T 3D facial reconstruction system from skull for Vietnamese %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2652/ %0 Conference Paper %A Ma, Thi Chau %A Nguyen, Thanh Thuy %B International Conference on Digital Arts, Media and Technology. %C Chieng Rai, Thailand %D 2018 %F SisLab:3044 %T A Labanotation based ontology for representing Vietnamese folk dances %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3044/ %X This paper aims at designing and constructing an ontology, representing and allowing Vietnamese folk dances annotation. An effective approach proposed is based on movement analysis. It is possible to divide Vietnamese folk dances into basic units, then movement phrases, movement primitives and dance poses. Studying deeply Vietnamese folk dances features and characteristics allows to find out 5 different types of relationships between main dance components. Typical relationships could be written in the form of logical predicates. For enriching ontologies, a formal encoding of the ontology expertise knowledge in the domain is performed by DL-Lite. The paper uses OWL-2 for building, organizing and visualizing ontologies. A case study – Mo folk dance with a knowledge base, is structured simply but effectively, with reasoning in DL-Lite and answering queries in SPARQL. %0 Conference Paper %A Ma Thi, Chau %A Park, Chang-soo %A Suthunyatanakit, Kittichai %A Oh, Min-jae %A Kim, Tae-wan %A Kang, Myung-joo %A Bui, The Duy %B Multimedia, Computer Graphics and Broadcasting %D 2012 %F SisLab:2651 %P 89-101 %T Features Detection from Industrial Noisy 3D CT Data for Reverse Engineering %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2651/ %X To detect features are significantly important for reconstructing a model in reverse engineering. In general, it is too difficult to find the features from the original industrial 3D CT data because the data have many noises. So it is necessary to reduce the noises for detecting features. This paper proposes a new method for detecting corner features and edge features from noisy 3D CT scanned data. First, we applied the level set method[18] to CT scanned image in order to segment the data. Next, in order to reduce noises, we exploited nonlocal means method[19] to the segmented surface. This helps to detect the edges and corners more accurately. Finally, corners and sharp edges are detected and extracted from the boundary of the shape. The corners are detected based on Sobel-like mask convolution processing with a marching cube. The sharp edges are detected based on Canny-like mask convolution with SUSAN method[13], which is for noises removal. In the paper, the result of detecting both features is presented. %0 Conference Paper %A Ma Thi, Chau %A Sylvain, Lagrue %A Nathalie, Chetcuti-Sperandio %A Ngo, Thi Duyen %A Fabien, Delorme %A Karim, Tabia %A Salem, Benferhat %B The 1st workshop on Structuring and Understanding of Multimedia heritAge Contents %C Nice, France %D 2019 %F SisLab:3750 %T An Ontology Web Application-based Annotation Tool for Intangible Culture Heritage Dance Videos %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3750/ %X Collecting dance videos, preserving and promoting them after enriching the collected data has been significant actions in preserving Intangible culture heritage in South-East Asia. Whereas techniques for the conceptual modeling of the expressive semantics of dance videos are very complex, they are crucial to exploit effectively the video semantics. This paper proposes an ontology web-based dance video annotation system for representing the semantics of dance videos at different granularity levels. Especially, the system incorporates both syntactic and semantic features of pre-built dance ontology system in order to not only use the available semantic web system but also to create unity for users when annotating videos to minimize conflicts. %0 Conference Paper %A Ma Thi, Chau %A Tabia, Karim %A Sylvain, Lagrue %A Nguyen, Thanh Thuy %A Le, Thanh Ha %A Bui, The Duy %B International Conference on Industrial, Engineering and Other Applications of Applied Intelligent Systems IEA/AIE 2017: Advances in Artificial Intelligence: From Theory to Practice %C Arras, France %D 2017 %F SisLab:2647 %P 3-11 %T Annotating movement phrases in Vietnamese folk dance videos %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2647/ %X This paper aims at the annotation of movement phrases in Vietnamese folk dance videos that were mainly gathered, stored and used in teaching at art schools and in preserving cultural intangible heritages (performed by different famous folk dance masters). We propose a framework of automatic movement phrase annotation, in which the motion vectors are used as movement phrase features. Movement phrase classification can be carried out, based on dancer’s trajectories. A deep investigation of Vietnamese folk dance gives an idea of using optical flow as movement phrase features in movement phrase detection and classification. For the richness and usefulness in annotation of Vietnamese folk dance, a lookup table of movement phrase descriptions is defined. In initial experiments, a sample movement phrase dataset is built up to train k-NN classification model. Experiments have shown the effectiveness of the proposed framework of automatic movement phrase annotation with classification accuracy at least 88%. %0 Conference Paper %A Mac, Hieu %A Truong, Dung %A Nguyen, Lam %A Nguyen, Ngoc Hoa %A Tran, Hai Anh %A Tran, Quang Duc %B The Ninth International Symposium on Information and Communication Technology (SoICT 2018) %C Da Nang %D 2018 %F SisLab:3166 %T Detecting Atacks on Web Applications using Autoencoder %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3166/ %X Web attacks have become a real threat to the Internet. This paper proposes the use of autoencoder to detect malicious pattern in the HTTP/HTTPS requests. The autoencoder is able to operate on the raw data and thus, does not require the hand-crafted features to be extracted. We evaluate the original autoencoder and its variants and end up with the Regularized Deep Autoencoder, which can achieve an F1-score of 0.9463 on the CSIC 2010 dataset. It also produces a better performance with respect to OWASP Core Rule Set and other one-class methods, reported in the literature. The Regularized Deep Autoencoder is then combined with Modsecurity in order to protect a website in real time. This algorithm proves to be comparable to the original Modsecurity in terms of computation time and is ready to be deployed in practice. %0 Journal Article %@ ISSN 2278-3091 %A Mai, Manh Trung %A Le, Phe Do %A Le, Trung Thuc %A Dao, Thi Phuong Anh %D 2020 %F SisLab:4028 %J International Journal of Advanced Trends in Computer Science and Engineering %N 3 %P 4158-4162 %T Proposing an Elliptic Curve Cryptosystem with the Symmetric Key for Vietnamese Text Encryption and Decryption %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4028/ %V 9 %X The article describes the basic idea of Elliptic curve cryptography (ECC). Elliptic curve arithmetic can be used to develop Elliptic curve coding schemes, including key exchange, encryption, and digital signature. The main attraction of Elliptic curve cryptography compared to RSA is that it provides equivalent security for a smaller key size, which reduces processing costs. To encode the Vietnamese text, we are based on the sound of Vietnamese characters to make a table of these characters’ order. We are also based on the algorithm to create the data sequence as the basis of building an encryption algorithm by using Elliptic curves on finite fields with symmetric keys to encrypt this Vietnamese text %0 Conference Paper %A Mai, Manh Trung %A Le, Phe Do %A Le, Trung Thuc %A Tran, Van Manh %A Le, Thi Len %A Nguyen, Khac Hung %A Nguyen, Thi Hang %B Một số vấn đề chọn lọc về an toàn thông tin %D 2017 %F SisLab:2754 %T Nghiên cứu các cuộc tấn công hệ mật mã nhẹ PRESENT %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2754/ %X PRESENT là một trong những họ mật mã khối hạng nhẹ có kích thước khối tin 64-bit với hai biến thể của khóa lần lượt là 64-bit và 128- bit. Thiết kế của PRESENT sử dụng nhiều các ánh xạ tuyến tính và sử dụng khóa trực tiếp trong khi mã hóa mà không thông qua những phép biến đổi làm việc trước với khóa. Do đó, PRESENT dường như khá mong manh trước những dạng tấn công thám mã như: tấn công đại số, tấn công khôi phục khóa... Trong bài báo này, chúng tôi thực hiện để xuất nâng cao thuật toán PRESENT bằng cách nâng cao S-Box nhằm mục đích nâng cao độ an toàn của thuật toán trước những nhà thám mã. Keywords-Mật mã nhẹ, mã khối, PRESENT Lightweight, Attack Present %0 Conference Paper %A Makino, Yoshiki %A Wongpatikaseree, Konlakorn %A Okada, Takashi %A Nguyen, Hoai Son %A Lim, Yuto %A Tan, Yasuo %B 7th International Conference on Information Communication Technology for Embedded Systems 2016 (IC-ICTES 2016) %D 2016 %F SisLab:1975 %P 45-50 %T Development of home simulation with thermal environment and electricity consumption %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1975/ %X Energy management systems have been developed in order to achieve goal of energy efficiency in the focused areas. Home simulation is one of the tools that can simulate electricity consumption based on the home appliances usage. However, several elements were ignored in traditional home simulation. In this paper, we introduce the home simulation, which can simulate not only the electric power and its consumption, but also the thermal energy consumption. Relationships among five elements are considered in the home simulation. To validate the performance of the proposed home simulation, a real experimental house is used to conduct the experiment. The results showed the ability of the proposed home simulation to simulate the thermal energy and estimate the temperature of various rooms with an error of 1.2 degrees Celsius. Furthermore, the electricity consumption is also simulated based on the human activity schedule and home appliances usage. %0 Conference Paper %A Man, Duc Chuc %A Luu, Viet Hung %A Hoang, Van Thang %A Bui, Quang Hung %A Nguyen, Thi Nhat Thanh %B 2014: The 6th International Conference on Knowledge and Systems Engineering (KSE 2014) %C Hanoi, Vietnam %D 2014 %F SisLab:443 %P 247-259 %T Cloud Detection Algorithm for LandSat 8 Image Using Multispectral Rules and Spatial Variability %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/443/ %0 Conference Paper %A Man, Duc Chuc %A Nguyen, Hoang Anh %A Nguyen, Thanh Thuy %A Bui, Quang Hung %A Nguyen, Thi Nhat Thanh %B The 9th International Conference on Knowledge and Systems Engineering (KSE 2017), 19-21 October 2017, Hue City, Vietnam. %D 2017 %F SisLab:2768 %T Paddy Rice Mapping in Red River Delta region Using Landsat 8 Images: Preliminary results %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2768/ %0 Journal Article %A McDonald, John F %A Mezencev, Roman %A Tran, Quoc Long %A Benigno, Benedict %A Bonta, Ioana %A Del Priore, Giuseppe %D 2015 %F SisLab:3275 %I American Society of Clinical Oncology %J Journal of Clinical Oncology %T Accurate prediction of optimal cancer drug therapies from molecular profiles by a machine-learning algorithm. %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3275/ %0 Journal Article %A Meysman, Pieter %A Dang, Thanh Hai %A Laukens, Kris %A De Smet, Riet %A Wu, Yan %A Marchal, Kathleen %A Engelen, Kristof %D 2011 %F SisLab:2075 %I Oxford Univ Press %J Nucleic Acids Research %N 2 %P e6 %T Use of structural DNA properties for the prediction of transcription-factor binding sites in Escherichia coli %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2075/ %V 39 %0 Journal Article %@ 1063-5157 %A Minh, Bui Quang %A Dang, Cao Cuong %A Vinh, Le Sy %A Lanfear, Robert %D 2021 %F SisLab:4533 %I Oxford University Press %J Systematic Biology %T QMaker: Fast and Accurate Method to Estimate Empirical Models of Protein Evolution %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4533/ %X Amino acid substitution models play a crucial role in phylogenetic analyses. Maximum likelihood (ML) methods have been proposed to estimate amino acid substitution models; however, they are typically complicated and slow. In this article, we propose QMaker, a new ML method to estimate a general time-reversible Q matrix from a large protein data set consisting of multiple sequence alignments. QMaker combines an efficient ML tree search algorithm, a model selection for handling the model heterogeneity among alignments, and the consideration of rate mixture models among sites. We provide QMaker as a user-friendly function in the IQ-TREE software package (http://www.iqtree.org) supporting the use of multiple CPU cores so that biologists can easily estimate amino acid substitution models from their own protein alignments. We used QMaker to estimate new empirical general amino acid substitution models from the current Pfam database as well as five clade-specific models for mammals, birds, insects, yeasts, and plants. Our results show that the new models considerably improve the fit between model and data and in some cases influence the inference of phylogenetic tree topologies.[Amino acid replacement matrices; amino acid substitution models; maximum likelihood estimation; phylogenetic inferences. %0 Journal Article %@ 1932-6203 %A Mokrousov, Igor %A Sintchenko, Vitali %A Anthony, Stephen %A Phan, Xuan Hieu %A Lin, Frank %A Coiera, Enrico W. %D 2010 %F SisLab:2294 %J PLoS ONE %N 3 %P e9535 %T A PubMed-Wide Associational Study of Infectious Diseases %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2294/ %V 5 %0 Conference Paper %A Monge, ME %A Zang, X %A Jones, CM %A Tran, Quoc Long %A Zhou, Manshui %A Walker L., DeEtte %A Mezencev, Roman %A Gray, Alexander %A McDonald, John F. %A Fernandez, Facundo %B 2nd Latin American Metabolic Profiling Symposium %D 2016 %F SisLab:2405 %T High Accuracy Prostate Cancer Detection Using Human Blood Serum Metabolomic Profiling %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2405/ %X Prostate cancer (PCa) represents the second leading cause of cancer mortality in many western countries. Although the Prostate-Specific Antigen (PSA) test is widely used to screen for PCa, certain advisory groups recommend against its use because it suffers from false positive results and over-treatment. These drawbacks have led to the increased interest of using metabolite profiling to discover new differential biomarkers that could improve the specificity of PCa diagnosis.1 In this work,2 untargeted metabolomic profiling of age-matched serum samples from PCa patients and healthy individuals was performed using ultraperformance liquid chromatography coupled to high-resolution tandem mass spectrometry and machine learning methods. PCa was detected in serum samples with 92.1% sensitivity, 94.3% specificity, and 93.0% accuracy by means of a metabolite-based in vitro diagnostic multivariate index assay, which outperformed the prevalent PSA test. Within the panel of 40 metabolic spectral features that was found to be differential, 31 metabolites were identified by MS and MS/MS, with 10 further confirmed by standards. Numerous discriminant metabolites were mapped in the steroid hormone biosynthesis pathway. The identification of fatty acids, amino acids, lysophospholipids, and bile acids provided further insights into the metabolite alterations associated with the disease. Our current work involves the prospective analysis of a larger sample cohort (n=500) that includes samples from PCa patients before and after undergoing surgery, patients with benign prostatitis, and healthy individuals with measured PSA values, in order to evaluate the influence of ethnicity and cohort size on the robustness of the metabolic biomarker panel that we have previously found, and to discover biomarkers useful for follow-up care. %0 Journal Article %@ 1863-2122 %A Mustafa, Tanveer %A Sohr, Karsten %A Dang, Duc Hanh %A Drouineaud, Michael %A Kowski, Stefan %D 2008 %F SisLab:1963 %J ECEASST %T Implementing Advanced RBAC Administration Functionality with USE %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1963/ %V 15 %0 Journal Article %@ 1813-9663 %A NGUYEN, Kim Sao %A Nguyen, Ngoc Hoa %A PHAM, Van At %D 2020 %F SisLab:3989 %J Journal of Computer Science and Cybernetics %N 2 %P 139-158 %T An effective reversible data hiding method based on pixel-value-ordering %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3989/ %V 36 %X This paper presents a new effective reversible data hiding method based on pixel-value-ordering (iGePVO-K) which is improvement of a recent GePVO-K method that recently is considered as a PVO-used method having highest embedding capacity. In comparison with GePVO-K method, iGePVO-K has the following advantages. First, the embedding capacity of the new method is higher than that of GePVO-K method by using data embedding formulas reasonably and reducing the location map size. Second, for embedding data, in the new method, each pixel value is modified at most by one, while in GePVO-K method, each pixel value may be modified by two. In fact, in the GePVO-K method, the largest pixels are modified by two for embedding bits 1 and by one for bits 0. This is also true for the smallest pixels. Meanwhile, in the proposed method, the largest pixels are modified by one for embedding bits 1 and are unchanged if embedding bits 0. Therefore, the stego-image quality in proposed method is better than that in GePVO-K method. Theoretical analysis and experiment results show that the proposed method has higher embedding capacity and better stego image quality than GePVO-K method. %0 Journal Article %@ 1949-2553 %A Naulaerts, Stefan %A Dang, Cao Cuong %A Ballester, Pedro %D 2017 %F SisLab:2582 %I Impact journals %J Oncotarget %T Precision and recall oncology: combining multiple gene mutations for improved identification of drug-sensitive tumours %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2582/ %X Cancer drug therapies are only effective in a small proportion of patients. To make things worse, our ability to identify these responsive patients before administering a treatment is generally very limited. The recent arrival of large-scale pharmacogenomic data sets, which measure the sensitivity of molecularly profiled cancer cell lines to a panel of drugs, has boosted research on the discovery of drug sensitivity markers. However, no systematic comparison of widely-used single-gene markers with multi-gene machine-learning markers exploiting genomic data has been so far conducted. We therefore assessed the performance offered by these two types of models in discriminating between sensitive and resistant cell lines to a given drug. This was carried out for each of 127 considered drugs using genomic data characterising the cell lines. We found that the proportion of cell lines predicted to be sensitive that are actually sensitive (precision) varies strongly with the drug and type of model used. Furthermore, the proportion of sensitive cell lines that are correctly predicted as sensitive (recall) of the best single-gene marker was lower than that of the multi-gene marker in 118 of the 127 tested drugs. We conclude that single-gene markers are only able to identify those drug-sensitive cell lines with the considered actionable mutation, unlike multi-gene markers that can in principle combine multiple gene mutations to identify additional sensitive cell lines. We also found that cell line sensitivities to some drugs (e.g. Temsirolimus, 17-AAG or Methotrexate) are better predicted by these machine-learning models. %0 Conference Paper %A Ngo, Kien Tuan %A Nguyen, Thu Trang %A Nguyen, Van Son %A Vo, Dinh Hieu %B 25th ACM International Systems and Software Product Line Conference (SPLC 2021) %C Leicester, UK %F SisLab:4523 %T Variability Fault Localization: A Benchmark %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4523/ %X Software fault localization is one of the most expensive, tedious, and time-consuming activities in program debugging. This activity becomes even much more challenging in Software Product Line (SPL) systems due to the variability of failures in SPL systems. These unexpected behaviors are caused by variability faults which can only be exposed under some combinations of system features. Although localizing bugs in non-configurable code has been investigated in-depth, variability fault localization in SPL systems still remains mostly unexplored. To approach this challenge, we propose a benchmark for variability fault localization with a large set of 1,773 buggy versions of six SPL systems and baseline variability fault localization performance results. Our hope is to engage the community to propose new and better approaches to the problem of variability fault localization in SPL systems. %0 Journal Article %@ 2588-1086 %A Ngo, Kien Tuan %A Vo, Dinh Hieu %A Bui, Ngoc Thang %A Pham, Le Viet Anh %A Pham, Khanh Ly %A Phan, Hai %D 2020 %F SisLab:4077 %I VNU %J VNU Journal of Science: Computer Science and Communication Engineering %T On Rectifying the Mapping between Articles and Institutions in Bibliometric Databases %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4077/ %X Today, bibliometric databases are indispensable sources for researchers and research institutions. The main role of these databases is to find research articles and estimate the performance of researchers and institutions. Regarding the evaluation of the research performance of an organization, the accuracy in determining institutions of authors of articles is decisive. However, current popular bibliometric databases such as Scopus and Web of Science have not addressed this point efficiently. To this end, we propose an approach to revise the authors’ affiliation information of articles in bibliometric databases. We build a model to classify articles to institutions with high accuracy by assembling the bag of words and n-grams techniques for extracting features of affiliation strings. After that, these features are weighted to determine their importance to each institution. Affiliation strings of articles are transformed into the new feature space by integrating weights of features and local characteristics of words and phrases contributing to the sequences. Finally, on the feature space, the support vector classifier method is applied to learn a predictive model. Our experimental result shows that the proposed model’s accuracy is about 99.1%. %0 Conference Paper %A Ngo, Le Minh %B Japan-Vietnam Workshop on Software Engineering %C Hanoi, Vietnam %D 2006 %F SisLab:2066 %T Towards building up an m-World %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2066/ %X The rapid development ofmobile computing has dramatically changed the way users view and use computers. With mobile computing technologies, we can acquire, deliver and process information at every place. Appearing in many forms such as the PalmOS, Pocket PC and Mobile Phone with increasingly diverse functions, and equipped with advanced peripheral devices such as GPS, camera and wireless adapter, mobile devices is becoming a very powerful all-in-one Personal Digital Assistant (PDA). PDAs have increased the effectiveness, brought people closer and made people more dynamic and flexible. In this paper, we present our research on setting up an m-World with various activities on PDAs. Our research focuses on three categories: education activities (mlearning), commerce activities (m-commerce) and service activities (m-service). In the first category, we have built a PDA-based m-learning system which allows the students to access the learning resources and allows the students and lecturers to exchange ideas from anywhere and at any time. In the category of commerce activities, we have developed experimental system which allows the user to do his/her shopping with mobile phones. There are various of systems which we have developed in the category of service activities including tourist assistant and traffic jam warning system. Our initial results have shown that such an m-World we are looking forward to is not very far from real. %0 Conference Paper %A Ngo, Le Minh %B International Conference on Computer Sciences dedicated to Research, Innovation and Vision for the Future %C Hanoi, Vietnam %D 2006 %F SisLab:2064 %T The impact of the implosion problem on the backoff time for a reliable multicast protocol %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2064/ %X This paper studies the implosion problem in multicast. The result explained in this paper is a part of the work on analyzing the scalability of a reliable multicast protocol. The paper describes a method proposed to determine the backoff time for multicast clients to avoid packet loss due to the implosion problem. The proposed method has been implemented and the output data is shown. The flow of acknowledgment packets from clients to the server has been studied. Random multicast networks are generated and backoff times are calculated in order to show the impact of the implosion problem on the transmission delay in reliable multicast. It is proved by analysis that the backoff interval is proportional to the number of users which is large enough. Linear increase in transmission delay is equivalent to that a reliable multicast protocol fails to work for a large number of users. %0 Conference Paper %A Ngo, Minh Hoang %A Nguyen, Xuan Viet Cuong %A Duong, Quang Khai %A Nguyen, Hoai Son %B 2019 25th Asia-Pacific Conference on Communications (APCC) %C Ho Chi Minh city, Vietnam %D 2019 %F SisLab:3607 %P 320-325 %T Adaptive Smart Lighting Control based on Genetic Algorithm %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3607/ %X Smart lighting system plays an important role in smart homes because of its convenience comparing with boring and tedious on/off switches. With conventional smart lighting systems, users can easily change the lighting level of a room with a smart phone or voice control device. However, design an automated lighting system with energy saving is still a challenge. In this paper, we propose an adaptive smart lighting system using a control algorithm based on Genetic Algorithm (GA). Our system turns on and off lights that match the user’s desire brightness and utilizes nature illuminance efficiently to save as much electric energy as possible. The control algorithm based on GA provides the optimized light turning pattern to the system and can adapt with the change of lighting environment and user requirement. Our evaluation results showed that the execution time of our algorithm is short enough to be used in real environment. %0 Thesis %9 PhD %A Ngo, Thi Duyen %B Khoa CNTT %D 2016 %F SisLab:1984 %I Trường ĐH Công nghệ - ĐHQGHN %T Nghiên cứu mô hình nhân vật ảo biểu cảm trên khuôn mặt ba chiều nói tiếng Việt %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1984/ %0 Conference Paper %A Ngo, Thi Duyen %A Akagi, Masato %A Bui, The Duy %B KSE: the 6th International Conference on Knowledge and Systems Engineering %C Hanoi, Vietnam %D 2014 %F SisLab:1416 %T Towards a Rule-Based Synthesis of Vietnamese Emotional Speech %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1416/ %0 Conference Paper %A Ngo, Thi Duyen %A Bui, The Duy %A VNU-UET, %B 2012 Fourth International Conference on Knowledge and Systems Engineering (KSE) %C Danang, Vietnam %D 2012 %F SisLab:99 %K Acoustics;Compounds;Databases;Feature extraction;Humans;Speech;Stress;acoustic signal processing;emotion recognition;speech synthesis;Vietnamese emotional speech prosody;Vietnamese sentence;Vietnamese synthesized speech;acoustic feature;cold anger;emotional style;happiness;hot anger;multistyle emotional speech database;neutral baseline;prosodic difference;prosodic variation;sadness;Acoustic Feature;Emotional Speech;Prosody;Vietnamese; %P 151 -155 %T A Study on Prosody of Vietnamese Emotional Speech %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/99/ %X This paper describes the analyses of the prosody of Vietnamese emotional speech, accomplished to find the relations between prosodic variations and emotional states in Vietnamese speech. These relations were obtained by investigating the variations of prosodic features in Vietnamese emotional speech in comparison with prosodic features of neutral speech. The analyses were performed on a multi-style emotional speech database which consisted of Vietnamese sentences uttered in different styles. Specifically, four emotional styles were considered: happiness, sadness, cold anger, and hot anger. Speech data in the neutral style were also collected, and prosodic differences of each style with respect to this neutral baseline were quantified. The acoustic features related to prosody which were investigated were fundamental frequency, power, and duration. According to the analysis results, for each speaker of the database, a set of prosodic variation coefficients was produced for each emotional style. This will help for bringing emotions into Vietnamese synthesized speech, making them more natural. %0 Conference Paper %A Ngo, Thi Duyen %A Bui, The Duy %B RIVF: 2015 IEEE RIVF International Conference on Computing Communication Technologies - Research, Innovation, and Vision for the Future %C Can Tho, Vietnam %D 2015 %F SisLab:1462 %K emotion recognition;human computer interaction;multi-agent systems;natural language processing;signal representation;speech synthesis;sport;Vietnamese 3D talking face;Vietnamese emotional speech synthesis module;combination module;communicative channels;embodied conversational agents;emotion-facial expressions module;emotional facial expressions;football supporter agent;football supporter domain;human computer interaction research societies;lip movement representation;multiagent;personality;three dimensional face;Acoustics;Databases;Face;Feature extraction;Lips;Speech;Three-dimensional displays;Conversational Agent;Emotional Facial Expression;Emotional Speech;Vietnamese 3D Talking Face %P 94-99 %T A Vietnamese 3D Taking Face for Embodied Conversational Agents %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1462/ %X Conversational agents are receiving significant attention from multi-agent and human computer interaction research societies. Many techniques have been developed to enable these agents to behave in a human-like manner. In order to do so, they are simulated with similar communicative channels as humans. Moreover, they are also simulated with emotion and personality. In this work, we focus on issue of expressing emotions for embodied-agents. We present a three dimensional face with ability to speak emotional Vietnamese speech and naturally express emotions while talking. Our face can represent lip movements during emotionally pronouncing Vietnamese words, and at the same time it can show emotional facial expressions while speaking. The face's architecture consists of three parts: Vietnamese Emotional Speech Synthesis module, Emotions to Facial Expressions module, and Combination module which creates lip movements when pronouncing Vietnamese emotional speech and combines these movements with emotional facial expressions. We have tested the face in the football supporter domain in order to confirm its naturalness. The face is simulated as the face of a football supporter agent which experiences emotions and expresses emotional expressions in his voice as well as on his face. %0 Conference Paper %A Ngo, Thi Duyen %A Bui, The Duy %B RIVF: the 2015 IEEE-RIVF International Conference on Computing and Communication Technologies %C Can Tho, Vietnam %D 2015 %F SisLab:1600 %P 94-99 %T A Vietnamese 3D Taking Face for Embodied Conversational Agents %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1600/ %X Conversational agents are receiving significant attention from multi-agent and human computer interaction research societies. Many techniques have been developed to enable these agents to behave in a human-like manner. In order to do so, they are simulated with similar communicative channels as humans. Moreover, they are also simulated with emotion and personality. In this work, we focus on issue of expressing emotions for embodied-agents. We present a three dimensional face with ability to speak emotional Vietnamese speech and naturally express emotions while talking. Our face can represent lip movements during emotionally pronouncing Vietnamese words, and at the same time it can show emotional facial expressions while speaking. The face's architecture consists of three parts: Vietnamese Emotional Speech Synthesis module, Emotions to Facial Expressions module, and Combination module which creates lip movements when pronouncing Vietnamese emotional speech and combines these movements with emotional facial expressions. We have tested the face in the football supporter domain in order to confirm its naturalness. The face is simulated as the face of a football supporter agent which experiences emotions and expresses emotional expressions in his voice as well as on his face. %0 Conference Paper %A Ngo, Thi Duyen %A Le, Thi Hoi %A Vu, Duy Loi %A Nguyen, Dinh Viet %B Hội thảo Khoa học Quốc gia lần thứ hai về Nghiên cứu, Phát triển và Ứng dụng Công nghệ Thông tin và Truyền thông (ICT.rda'04) %C Hà Nội, Việt Nam %D 2005 %F SisLab:2992 %P 394-403 %T ĐÁNH GIÁ HIỆU SUẤT CHIẾN LƯỢC QUẢN LÝ HÀNG ĐỢI RED BẰNG BỘ MÔ PHỎNG MẠNG NS %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2992/ %X Báo cáo này trình bày các kết quả nghiên cứu của chúng tôi về RED (Random Early Detection of congestion) bằng bộ mô phỏng mạng NS. Chúng tôi cũng thực hiện các mô phỏng hàng đợi kiểu Tail-Drop để so sánh với RED. Vì lưu lượng trên Internet có đặc tính bùng nổ nên hàng đợi (bộ đệm) tại các nút mạng (router) phải có kích thước đủ lớn, để đảm bảo cho các nút thực hiện chức năng store-and-forward một cách hiệu quả. Tuy nhiên, nếu thi hành chính sách phục vụ tại hàng đợi kiểu FIFO (Tail-Drop Queue) thì hàng đợi sẽ thường xuyên ở trạng thái đầy, làm tăng đáng kể thời gian trễ trung bình của các gói tin trong mạng. Do vậy, điều quan trọng là phải có các kỹ thuật để đảm bảo cho mạng đạt được thông lượng cao và thời gian trễ trung bình nhỏ. RED là một chiến lược quản lý hàng đợi động, cộng tác với chiến lược tránh tắc nghẽn tại nguồn của TCP; RED đáp ứng được các yêu cầu nêu trên. RED phát hiện sớm sự tắc nghẽn mạng ngay khi nó có dấu hiệu sắp xuất hiện và phản ứng một cách tích cực để ngăn chặn. %0 Conference Paper %A Ngo, Thi Duyen %A Ma, Thi Chau %A Bui, The Duy %B The 6th International Conference on Knowledge and Systems Engineering (KSE) %C Hanoi, Vietnam %D 2014 %F SisLab:444 %P 487-498 %T Emotional Facial Expression Analysis in the Time Domain %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/444/ %X Emotions have been studied for a long time and results show that they play an important role in human cognitive functions. In fact, emotions play an extremely important role during the communication between people. And the human face is the most communicative part of the body for expressing emotions; it is recognized that a link exists between facial activity and emotional states. In order to make computer applications more believable and friendly, giving them the ability to recognize and/or express emotions are research fields which have been much focused on. Being able to perform these tasks, firstly, we need to have knowledge about the relationship between emotion and facial activity. Up to now, there have been proposed researches on this relationship. However, almost all these researches focused on analyzing the relationship without taking into account time factors. They analyzed the relationship but did not examined it in the time domain. In this paper, we propose a work on analyzing the relationship between emotions and facial activity in the time domain. Our goal is finding the temporal patterns of facial activity of six basic emotions (happy, sad, angry, fear, surprise, disgust). To perform this task, we analyzed a spontaneous video database in order to consider how facial activities which are related to the six basic emotions happen temporally. From there, we bring out the general temporal patterns for facial expressions of each of the six emotions. %0 Journal Article %A Ngo, Thi Duyen %A Tran, Thi Minh %A Ma, Thi Chau %A Nguyen, Xuan Hung %A Le, Thanh Ha %D 2019 %F SisLab:3760 %J International Association for Printing Woodblocks (IAPW) %P 333-348 %T Royal printing woodblocks of Nguyen Dynasty: 3D reconstruction for digital preversation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3760/ %0 Conference Paper %A Ngo, Thi Duyen %A Vu, Thi Hong Nhan %A Nguyen, Viet Ha %A Bui, The Duy %B 2014: the 17th International Conference on Principles and Practice of Multi-Agent Systems (PRIMA) %C Queensland, Australia %D 2014 %F SisLab:1427 %P 222-237 %T Improving Simulation of Continuous Emotional Expressions by Analyzing Video of Human Facial Activities %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1427/ %0 Conference Paper %A Ngo, Thi Duyen %A Vu, Thi Hong Nhan %A Nguyen, Viet Ha %A Bui, The Duy %B PRIMA: the 17th International Conference on Principles and Practice of Multi-Agent Systems %C Queensland, Australia %D 2014 %F SisLab:1417 %T Improving Simulation of Continuous Emotional Facial Expressions by Analyzing Videos of Human Facial Activities %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1417/ %0 Conference Paper %A Ngo, Thi Lan %A Duong, Quang Vu %A Pham, Son Bao %A Phan, Xuan Hieu %B Proceedings of the Seventh Symposium on Information and Communication Technology (SoICT) %C Ho Chi Minh City, Viet Nam %D 2016 %F SisLab:2293 %P 396-402 %T Speech act classification in Vietnamese Utterance and its application in smart mobile voice interaction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2293/ %0 Conference Paper %A Ngo, Thi Lan %A Nguyen, Van Hop %A Vuong, Thi Hai Yen %A Nguyen, Thac Thong %A Nguyen, Thi Thua %A Pham, Bao Son %A Phan, Xuan Hieu %B ACIIDS 2016: the 8th Asian Conference on Intelligent Information and Database Systems %C Da Nang, Vietnam %D 2016 %F SisLab:1640 %P 190-201 %T Identifying User Intents in Vietnamese Spoken Language Commands and Its Application in Smart Mobile Voice Interaction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1640/ %V 9621 %0 Conference Paper %A Ngo, Thi Lan %A Pham, Bao Son %A Pham, Khac Linh %A Phan, Xuan Hieu %A Cao, Minh Son %B The 9th International Conference on Knowledge and Systems Engineering (KSE 2017) %C Hue city, Vietnam %D 2017 %F SisLab:2687 %P 203-208 %T Dialogue act segmentation for Vietnamese human-human conversational texts %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2687/ %X Dialog act identification plays an important role in understanding conversations. It has been widely applied in many fields such as dialogue systems, automatic machine translation, automatic speech recognition, and especially useful in systems with human-computer natural language dialogue interfaces such as virtual assistants and chatbots. The first step of identifying dialog act is identifying the boundary of the dialog act in utterances. In this paper, we focus on segmenting the utterance according to the dialog act boundaries, i.e. functional segments identification, for Vietnamese utterances. We investigate carefully functional segment identification in two approaches: (1) machine learning approach using maximum entropy (ME) and conditional random fields (CRFs); (2) deep learning approach using bidirectional Long Short-Term Memory (LSTM) with a CRF layer (Bi-LSTM-CRF) on two different conversational datasets: (1) Facebook messages (Message data); (2) transcription from phone conversations (Phone data). To the best of our knowledge, this is the first work that applies deep learning based approach to dialog act segmentation. As the results show, deep learning approach performs appreciably better as to compare with traditional machine learning approaches. Moreover, it is also the first study that tackles dialog act and functional segment identification for Vietnamese. %0 Conference Paper %A Ngo, Thi Lan %A Pham, Bao Son %A Phan, Xuan Hieu %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1529 %T Identifying User Intents in Vietnamese Spoken Language Commands and Its Application in Smart Mobile Voice Interaction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1529/ %X This paper presents a lightweight machine learning model and a fast conjunction matching method to the problem of identifying user intents behind their spoken text commands. These model and method were integrated into a mobile virtual assistant for Vietnamese (VAV) to understand what mobile users mean to carry out on their smartphones via their commands. User intent, in the scope of our work, is an action associated with a particular mobile application. Given an input spoken command, its application will be identified by an accurate classifier while the action will be determined by a flexible conjunction matching algorithm. Our classifier and conjunction matcher are very compact in order that we can store and execute them right on mobile devices. To evaluate the classifier and the matcher, we annotated a medium–sized data set, conducting various experiments with different settings, and achieving impressive accuracy for both the application and action identification. %0 Conference Paper %A Ngo, Thi Lan %A Pham, Khac Linh %A Hideaki, Takeda %A Pham, Bao Son %A Phan, Xuan Hieu %B The Eighth International Symposium on Information and Communication Technology (SoICT 2017) %C Nha Trang %D 2017 %F SisLab:2692 %T On the Identification of Suggestion Intents from Vietnamese Conversational Texts %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2692/ %X Suggestion intents mining from texts is an emerging research topic in recent years. Fully understanding suggestion intents in conversational texts is a complicated process that includes three major stages: user suggestion intents filtering, suggestion domain identification, and arguments extraction of suggestion intents. In the scope of this paper, we study the first phase, that is, building a binary classification model to determine whether a text unit carries suggestion intents or not. We come up with a new text unit to analysis suggestion, that is functional segment. According to the ISO 24617-2 standard, a functional segment is “minimal stretch of communicative behavior that has one or more communicative functions”. We investigate two approaches to filter functional segment containing suggestion intents: machine learning using maximum entropy (Maxent) model and deep learning using convolutional neural networks (CNN) model. The results of these experiments on Vietnamese online media texts are very promising. To the best of our knowledge, this is the first study to analyze suggestion at functional segment unit. %0 Conference Paper %A Ngo, Thi Lan %A Vu, Van Tu %A Hideaki, Takeda %A Pham, Bao Son %A Phan, Xuan Hieu %B 19th International Conference on Computational Linguistics and Intelligent Text Processing %C Hanoi, Vietnam %D 2018 %F SisLab:2980 %T Lifelong Learning MaxEnt for Suggestion Classification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2980/ %X Suggestion analysis of opinion data is classifying a given utterance into one of two classes: suggestion and non-suggestion. In this paper, we introduce a new method, called LLMaxent, to cross-domain suggestion classification. LLMaxent is an approach to lifelong machine learning using maximum entropy (Maxent) method. Based on the main idea of lifelong learning, that is retaining the knowledge learned from past tasks and using it to help future learning, we build a classifier can use labeled data in existed domains for suggestion classification in a new domain. The experimental results show that the proposed novel model can improve the performance of cross-domain suggestion classification. This is the first study of lifelong machine learning using Maxent and our method is not only useful for suggestion classification but also for cross-domain text classification in general. %0 Conference Paper %A Ngo, Thi Oanh %A Chu, Chi Quang %A Lai, Thi Huyen Trang %A Nguyen, Thi Dung %A Ha, Quang Thuy %A Le, Hoang Quynh %B SSC: the 2016 Student Scientific Contest %C Hanoi %D 2016 %F SisLab:1495 %T Xây dựng mô hình và hệ thống dự báo dịch tả khu vực Hà Nội dựa trên kỹ thuật khai phá dữ liệu %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1495/ %0 Conference Paper %A Ngo, Thi-Duyen %A Akagi, Masato %A Bui, The Duy %B 2014: The 6th International Conference on Knowledge and Systems Engineering (KSE 2014) %C Hanoi, Vietnam %D 2014 %F SisLab:440 %P 129-142 %T Toward a Rule-Based Synthesis of Vietnamese Emotional Speech %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/440/ %0 Journal Article %A Ngo, Truong X. %A Do, Ngoc T.N. %A Phan, Hieu D.T. %A Tran, Vinh T. %A Mac, Tra T.M. %A Le, Anh H. %A Do, Nguyet V. %A Bui, Hung Q. %A Nguyen, Thanh T.N. %D 2021 %F SisLab:4536 %I Taylor & Francis %J International Journal of Remote Sensing %N 16 %P 6128-6154 %T Air pollution in Vietnam during the COVID-19 social isolation, evidence of reduction in human activities %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4536/ %V 42 %0 Conference Paper %A Ngo, Xuan Bach %A Le, Anh Cuong %A Nguyen, Viet Ha %A Nguyen, Ngoc Binh %B Advanced Language Processing and Web Information Technology %D 2008 %F SisLab:1777 %P 9-14 %T Transformation Rule Learning without Rule Templates: A Case Study in Part of Speech Tagging %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1777/ %X Part of speech (POS) tagging is an important problem and is one of the first steps included in many tasks in natural language processing. It affects directly on the accuracy of many other problems such as Syntax Parsing, WordSense Disambiguation, and Machine Translation. Stochastic models solve this problem relatively well, but they still make mistakes. Transformation-based learning (TBL) is a solution which can be used to improve stochastic taggers by learning a set of transformation rules. However, its rule learning algorithm has the disadvantages that rule templates must be prepared by hand and only rules are instances of rule templates can be generated. In this paper, we propose a model to learn transformation rules without rule templates. This model considers the rule learning problem as a feature selection problem. Experiments on PennTree Bank showed that the proposal model reduces errors of stochastic taggers with some tags. %0 Conference Paper %A Ngo, Xuan Truong %A Nguyen, Van Hai %A Tran, Trung Kien %A Do, Thanh Cong %A Ha, Duc Van %A Luu, Quang Thang %A Luu, Viet Hung %A Bui, Quang Hung %A Nguyen, Thi Nhat Thanh %B Hội nghị Quốc gia lần thứ XXI về Điện tử, Truyền thông và Công nghệ Thông tin %C Hà Nội, Việt Nam %D 2018 %F SisLab:3325 %P 60-67 %T Thiết kế và chế tạo vệ tinh FIMO CANSAT %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3325/ %X Ngày nay, vệ tinh đã và đang vai trò quan trọng trong nhiều lĩnh vực cuộc sống. Vệ tinh có nhiều loại, nhiều kích thước được sử dụng cho cácmục đích khác nhau. CANSAT là một loại vệ tinh cỡ nhỏ có kích thước bằng lon nước được sử dụng cho mục đích giáo dục. Trong bài báo này, chúng tôi sẽ trình bày phương pháp thiết kế và chế tạo vệtinh FIMO CANSAT,quá trình kiểm thử và phóng vệ tinh.Vệtinh FIMO CANSAT được chế tạo với mục đích giám sát chất lượng không khí, một trong những vấn đề cấp thiết và đang được quan tâm hiện nay. Các cảm biến trong vệ tinh được kiểm thử trong các điều kiện môi trường khác nhau. Vệ tinh có khả năng đo nhiệt độ , độ ẩm, bụi mịn (PM1/2.5/10), CO và truyền thông tốt với trạm mặt đất trong khoảng cách 300m. Trong thực nghiệm, vệ tinh đã hoạt động ổn định từ lúc phóng tên lửa mô hình đến khi vệ tinh rơi chạm đất, dữ liệu thu được từ vệ tinh có độ chính xác tương đối, phản ánh sự thay đổi môi trường thực tế theo độ cao. Ngoài ra, phần mềm giám sát hoạt động CANSAT cũng sẽ được giới thiệu trong bài báo. %0 Book Section %A Nguyen, Ba Tung %A Do, Khac Phong %A Tran, Nguyen Le %A Bui, Quang Hung %A Nguyen, Thi Nhat Thanh %A Vuong, Van Quynh %A Le, Thanh Ha %B Land-Atmospheric Research Applications in South and Southeast Asia. Springer Remote Sensing/Photogrammetry %D 2018 %F SisLab:3288 %I Springer, Cham %T Enhancement of Fire Early Warning System in Vietnam Using Spatial Data and Assimilation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3288/ %0 Conference Paper %A Nguyen, Cam Tu %A Kaothanthong, Natsuda %A Phan, Xuan Hieu %A Tokuyama, Takeshi %B The 19th ACM Conference on Information and Knowledge Management (ACM CIKM) %C Toronto, Canada %D 2010 %F SisLab:2295 %P 1481-1484 %T A feature-word-topic model for image annotation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2295/ %0 Journal Article %@ 1559-1131 %A Nguyen, Cam Tu %A Kaothanthong, Natsuda %A Tokuyama, Takeshi %A Phan, Xuan Hieu %D 2013 %F SisLab:1889 %J ACM Transactions on the Web %N 3 %P 1-24 %T A feature-word-topic model for image annotation and retrieval %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1889/ %V 7 %X Image annotation is a process of finding appropriate semantic labels for images in order to obtain a more convenient way for indexing and searching images on the Web. This article proposes a novel method for image annotation based on combining feature-word distributions, which map from visual space to word space, and word-topic distributions, which form a structure to capture label relationships for annotation. We refer to this type of model as Feature-Word-Topic models. The introduction of topics allows us to efficiently take word associations, such as {ocean, fish, coral} or {desert, sand, cactus}, into account for image annotation. Unlike previous topic-based methods, we do not consider topics as joint distributions of words and visual features, but as distributions of words only. Feature-word distributions are utilized to define weights in computation of topic distributions for annotation. By doing so, topic models in text mining can be applied directly in our method. Our Feature-word-topic model, which exploits Gaussian Mixtures for feature-word distributions, and probabilistic Latent Semantic Analysis (pLSA) for word-topic distributions, shows that our method is able to obtain promising results in image annotation and retrieval. %0 Conference Paper %A Nguyen, Cam Tu %A Le, Vu Ha %A Tokuyama, Takeshi %B International Conference on Knowledge Discovery and Information Retrieval %C Paris, France %D 2011 %F SisLab:1848 %P 14-23 %T Cascade of Multi-level Multi-instance Classifiers for Image Annotation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1848/ %0 Conference Paper %A Nguyen, Cam Tu %A Nguyen, Trung Kien %A Phan, Xuan Hieu %A Nguyen, Le Minh %A Ha, Quang Thuy %B The 20st Pacific Asia Conference on Language, Information and Computation (PACLIC) %C Wuhan, China %D 2006 %F SisLab:2305 %P 215-222 %T Vietnamese Word Segmentation with CRFs and SVMs: An Investigation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2305/ %0 Journal Article %@ 15300226 %A Nguyen, Cam Tu %A Phan, Xuan Hieu %A Horiguchi, Susumu %A Nguyen, Thu Trang %A Ha, Quang Thuy %D 2009 %F SisLab:2296 %J ACM Transactions on Asian Language Information Processing (ACM TALIP) %N 3 %P 1-40 %T Web Search Clustering and Labeling with Hidden Topics %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2296/ %V 8 %0 Conference Paper %A Nguyen, Cam-Tu %A Wang, Xiaoliang %A Liu, Jing %A Zhou, Zhi-Hua %B 2014: the 28th Conference on Artificial Intelligence (AAAI) %C Québec, Canada %D 2014 %F SisLab:1302 %T Multi-View Multi-Instance Multi-Label Learning %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1302/ %0 Conference Paper %A Nguyen, Cam-Tu %A Zhan, D.-C. %A Zhou, Z.-H. %B 2013 the 23rd International Joint Conference on Artificial Intelligence (IJCAI'13) %C Beijing, China %D 2013 %F SisLab:190 %T Multi-Modal Image Annotation with Multi-Instance Multi-Label LDA %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/190/ %0 Report %9 Technical Report %A Nguyen, Cam-Van Thi %A Tran Mai-Vu, %D 2019 %F SisLab:3719 %T A Large Scale Multi-label Text Classification Methodusing Z-Label LDA %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3719/ %X Multi-label Learning (MLL) is a supervised learning model that hasattracted much attention of the research community in recent yearsbecause of its wide variety applicability. In this paper, we built amulti-label classification model using Latent Dirichlet Allocationwith Topic-in-set Knowlegde (z-Label LDA) on the Vietnamesedata domain. z-Label LDA is a variant of LDA which is intended toprovide additional supervised information as a hidden topic into theLDA called "z-label". We also have experimented on the dataset inthe field of Education collected from Vietnamese online newspapers.Parallel, we applied the hidden topic model LDA to generate aprior-knowledge dataset comprising topics and typical keywordsrepresenting each topic. The supervised information also makesthe topic assignment more consistent. With this approach, theeffectiveness of the model has been demonstrated experimentally,this paper has obtained initial positive results. %0 Conference Paper %A Nguyen, Cong Hung %A Nguyen, Quynh Mai %A Pham, Ngoc Hung %A Vo, Dinh Hieu %B SSC: the 2016 Student Scientific Contest %C Hanoi %D 2016 %F SisLab:1499 %T Phương pháp và công cụ phân tích ảnh hưởng của sự thay đổi mã nguồn cho các ứng dụng J2EE6 %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1499/ %0 Conference Paper %A Nguyen, Dai Thanh %A Pham, Thi Minh Trang %A Dang, Thanh HaI %A Nguyen, Ha Anh Tuan %A Le, Si Quang %A Bui, Quang Minh %A Dao, Quang Minh %A Pham, Bao Son %A Le, Sy Vinh %B KSE: the 2015 International Conference on Knowledge and Systems Engineering %C Ho Chi Minh city, Vietnam %D 2015 %F SisLab:1608 %P 97-102 %T Building population-specific reference genomes: A case study of Vietnamese reference genome %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1608/ %X The human reference genome is an essential tool for studying human genomes. The standard reference genome is constructed from genomes of a few donors. The 1000 genomes project has revealed a huge amount of genetic differences between diverse populations. It is therefore naturally questioned whether the standard reference genome can work well for all human genome studies or population-specific reference genomes are needed accordingly. In this paper, we present a pipeline for constructing and evaluating a population-specific reference genome. The pipeline was examined on building the Vietnamese reference genome from 100 Kinh Vietnamese genomes obtained from the 1000 genomes project. Experiments showed that the resulting Vietnamese reference genome was better than the standard reference genome at analyzing Vietnamese genomic data. It helped improve the quality of short reads mapping and genotype calling for Vietnamese genomes. The pipeline is applicable for building and evaluating other population-specific reference genomes. For the first time the Vietnamese reference genome, which is now available for further Vietnamese genome studies, was successfully built. %0 Journal Article %@ 2305-5839 %A Nguyen, Dai Thanh %A Pham, Thi Minh Trang %A Dang, Thanh Hai %A Nguyen, Ha Anh Tuan %A Le, Si Quang %A Bui, Quang Minh %A Dao, Quang Minh %A Pham, Bao Son %A Le, Sy Vinh %D 2015 %F SisLab:2083 %I AME Publications %J Annals of translational medicine %N Suppl %T AB050. Building population-specific reference genomes: a case study of Vietnamese reference genome %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2083/ %V 3 %0 Conference Paper %A Nguyen, Dai Tho %A Ho, Thuan %A Le, Dinh Thanh %A University of Engineering and Technology, Vietnam National University, Hanoi, %A Institute of Information Technology, Vietnam Academy of Science and Technology, %B Hội thảo quốc gia lần thứ XV: Một số vấn đề chọn lọc của Công nghệ thông tin và truyền thông %C Hanoi, Vietnam %D 2013 %F SisLab:2368 %P 122-126 %T Improvements to a Protocol for the Maintenance of Common Data in Distributed Systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2368/ %X In this paper, we suggest some modifications to a protocol proposed by Awerbuch and Schulman in [3] for the maintenance of common data in distributed systems so that both its time and communication complexities are reduced. The first change that makes Awerbuch-Schulman protocol better is to eliminate unnecessary messages used in the original version. The second change that also improves the protocol is to reduce the size of process messages. In addition, we suggest a self-stabilizing version of the Awerbuch-Schulman protocol as well. %0 Conference Paper %A Nguyen, Dai Tho %A Le, Dinh Thanh %A Nguyen, Binh Minh %B ICCSAMA: the 2nd International Conference on Computer Science, Applied Mathematics and Applications %C Budapest, Hungary %D 2014 %F SisLab:1414 %P 415-426 %T Efficient Core Selection for Multicast Routing in Mobile Ad Hoc Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1414/ %0 Journal Article %@ 1471-2105 %A Nguyen, Dat Thanh %A Trac, Quang Thinh %A Nguyen, Thi-Hau %A Nguyen, Ha-Nam %A Ohad, Nir %A Pawitan, Yudi %A Vu, Trung Nghia %D 2021 %F SisLab:4647 %J BMC Bioinformatics %N 1 %T Circall: fast and accurate methodology for discovery of circular RNAs from paired-end RNA-sequencing data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4647/ %V 22 %0 Conference Paper %A Nguyen, Diep Thi Ngoc %B ICCV 2019 - The international Workshop and Challenge on Real-world Recognition from Low-quality images and videos %C Seoul, Republic of Korea %D 2019 %F SisLab:3813 %T State-of-the-Art in Action: Unconstrained Text Detection %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3813/ %X In this paper, we stage five real-world scenarios for six state-of-the-art text detection methods in order to evaluate how competent they are with new data without any training process. Moreover, this paper analyzes the architecture design of those methods to reveal the influence of pipeline choices on the detection quality. The setup of experimental studies are straight-forward: we collect and manually annotate test data, we reimplement the pretrained models of the state-of-the-art methods, then we evaluate and analyze how well each method achieve in each of our collected datasets. We found that most of the state-of-the-art methods are competent at detecting textual information in unseen data, however, some are more readily used for real-world applications. Surprisingly, we also found that the choice of a post-processing algorithm correlates strongly with the performance of the corresponding method. We expect this paper would serve as a reference for researchers as well as application developers in the field. All collected data with ground truth annotation and their detected results is publicly available at our Github repository: https://github.com/chupibk/ HBlab-rlq19. %0 Journal Article %A Nguyen, Diep Thi Ngoc %A Kiyoki, Yasushi %D 2014 %F SisLab:3816 %I IOS Press %J Frontiers in Artificial Intelligence and Applications %P 324-342 %T An Adaptive Search Path Traverse for Large-scale Video Frame Retrieval %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3816/ %V 272 %X Multimedia retrieval task is faced with increasingly large datasets and variously changing preferences of users in every query. We realize that the high dimensional representation of physical data which previously challenges search algorithms now brings chances to cope with dynamic contexts. In this paper, we introduce a method of building a large-scale video frame retrieval environment with a fast search algorithm that handles user's dynamic contexts of querying by imagination and controlling response time. The search algorithm quickly finds an initial candidate, which has highest-match possibility, and then iteratively traverses along feature indexes to find other neighbor candidates until the input time bound is elapsed. The experimental studies based on the video frame retrieval system show the feasibility and effectiveness of our proposed search algorithm that can return results in a fraction of a second with a high success rate and small deviation to the expected ones. Moreover, its potential is clear that it can scale to large dataset while preserving its search performance. %0 Conference Paper %A Nguyen, Diep Thi Ngoc %A Kiyoki, Yasushi %B 2016 International Electronics Symposium %C Denpasar, Indonesia %D 2016 %F SisLab:3814 %P 357-362 %T A bigraph representation model and directional search mechanism for debates %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3814/ %X The beauty of thinking resides not only in finding answers but also in challenging already existing ones and creating intelligible networks of thoughts and ideas of many different perspectives. Understanding that the essence of critical thinking presents in the flows of questions and answers and aiming at a tool of augmenting it, we propose a new knowledge representation model based on bipartite graphs and a set of functions aided to explore those graphs. A bipartite graph has two disjoint vertex sets that are question and answer sets. Every edge connects a question to an answer holds an evidence for the answer and every edge connects an answer to a question holds an argument raised from the answer. The directional search mechanism is particularly designed to reuse reasoning flows in the debate graphs by a projecting function between the question and answer sets. We also introduce a framework placing the model and the search mechanism altogether in collaborative applications. Many philosophical debates are collected to demonstrate the advantages of the bigraph model besides its simplicity. The analysis and experiments on optimizing semantic search show a sufficient performance for a real-time application. %0 Journal Article %A Nguyen, Diep Thi Ngoc %A Kiyoki, Yasushi %D 2014 %F SisLab:3815 %I IOS press %J Frontiers in Artificial Intelligence and Applications %P 324-342 %T http://ebooks.iospress.nl/publication/38509 %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3815/ %V 272 %X Multimedia retrieval task is faced with increasingly large datasets and variously changing preferences of users in every query. We realize that the high dimensional representation of physical data which previously challenges search algorithms now brings chances to cope with dynamic contexts. In this paper, we introduce a method of building a large-scale video frame retrieval environment with a fast search algorithm that handles user's dynamic contexts of querying by imagination and controlling response time. The search algorithm quickly finds an initial candidate, which has highest-match possibility, and then iteratively traverses along feature indexes to find other neighbor candidates until the input time bound is elapsed. The experimental studies based on the video frame retrieval system show the feasibility and effectiveness of our proposed search algorithm that can return results in a fraction of a second with a high success rate and small deviation to the expected ones. Moreover, its potential is clear that it can scale to large dataset while preserving its search performance. %0 Journal Article %A Nguyen, Diep Thi Ngoc %A Kiyoki, Yasushi %D 2013 %F SisLab:3819 %J Frontiers in Artificial Intelligence and Applications %P 201-220 %T An imagination-based query creation method for image retrieval %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3819/ %V 251 %X Our imagination-based query creation method is a new approach to image-recall functions reflecting color-based imaginations in human brains. This method dynamically represents user's imaginations and creates queries for image retrieval. The important aim of this method is to express user's abstract intentions in a computable objects and to recall those imaginations from image databases with color-analytical image retrieval. The main features of this method are: (1) a hierarchical model for color indexing to express image contexts, (2) color zooming for context-dependent color histogram generation by utilizing color names as contextual words, (3) integrating several histograms by using five histogram-combining operations and dynamic semantic weighting, and (4) threshold controlling for semantic correlation. This method exploits color information to retrieve intended images expressing a user's imagination. This paper also represents a “Database for Imagination Expression” (DBIE), which stores, shares and reuses created queries. Additionally, several qualitative experiments are also represented to examine the effectiveness and the feasibility of our method. %0 Conference Paper %A Nguyen, Diep Thi Ngoc %A Nakayama, Hideki %A Okazaki, Naoaki %A Sakaeda, Tatsuya %B The 26th ACM international conference on Multimedia %C Seoul, Korea %D 2018 %F SisLab:3788 %P 1786-1793 %T PoB %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3788/ %X Aiming to develop of computational grammar system for visual information, we design a 4-tier framework that consists of four levels of 'visual grammar of images.' As a first step of realization, we propose a new dataset, named the PoB dataset, in which each image is annotated with multiple labels of armature patterns that compose the pictorial scene. The PoB dataset includes of a 10,000-painting dataset for art and a 4,959-image dataset for photography. In this paper, we discuss the consistency analysis of our dataset and its applicability. We also demonstrate how the armature patterns in the PoB dataset are useful in assessing aesthetic quality of images, and how well a deep learning algorithm can recognize these patterns. This paper seeks to set a new direction in image understanding with a more holistic approach beyond discrete objects and in aesthetic reasoning with a more interpretative way. %0 Conference Paper %A Nguyen, Diep Thi Ngoc %A Nakayama, Hideki %A Okazaki, Naoaki %A Sakaeda, Tatsuya %B The 26th ACM international conference on Multimedia %C Seoul, Republic of Korea %D 2018 %F SisLab:3790 %P 1786-1793 %T PoB: Toward Reasoning Patterns of Beauty in Image Data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3790/ %X Aiming to develop of computational grammar system for visual information, we design a 4-tier framework that consists of four levels of 'visual grammar of images.' As a first step of realization, we propose a new dataset, named the PoB dataset, in which each image is annotated with multiple labels of armature patterns that compose the pictorial scene. The PoB dataset includes of a 10,000-painting dataset for art and a 4,959-image dataset for photography. In this paper, we discuss the consistency analysis of our dataset and its applicability. We also demonstrate how the armature patterns in the PoB dataset are useful in assessing aesthetic quality of images, and how well a deep learning algorithm can recognize these patterns. This paper seeks to set a new direction in image understanding with a more holistic approach beyond discrete objects and in aesthetic reasoning with a more interpretative way. %0 Conference Paper %A Nguyen, Diep Thi Ngoc %A Sasaki, Shiori %A Kiyoki, Sasushi %B IASTED Multiconferences - Proceedings of the IASTED International Conference on Web-Based Education, WBE 2013 %C Innsbruck, Austria %D 2013 %F SisLab:3820 %P 909-916 %T WEE: A Mashup System Using Social Networks For Collaborative Environmental Monitoring Education With Spatiotemporal Analysis And Visualization %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3820/ %X The aim of this research is to construct a new collaborative mashup system, named Web-based Environmental Education (WEE), for global environmental monitoring research and education in which all participants could associatively organize multimedia data about physical environments and visually monitor various environmental topics based on time-series multi-geographical views. The main features of this system are: (1) collaborative multimedia database creation using "social sensors", (2) information aggregation from Web resources, (3) content-based image and document retrieval, and (4) big-picture visualization with spatiotemporal view controllers and content-based image, document retrieval. In this paper, a new designed usage of social network services as an educational tool and a new method for gathering environmental information from Web resources are presented. We also present three promising educational scenarios to examine the applicability of the system. %0 Conference Paper %A Nguyen, Diep Thi Ngoc %A Sasaki, Shiori %A Kiyoki, Yasushi %B IADIS International Conference (e-Society 2011) %C Spain %D 2011 %F SisLab:3824 %P 272-278 %T 5DPicMap: Imagination-based I mage Search System with Spatiotemporal Analyzers %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3824/ %0 Conference Paper %A Nguyen, Diep Thi Ngoc %A Sasaki, Shiori %A Kiyoki, Yasushi %B 13th IASTED International Conference on Software Engineering and Applications (SEA 2010) %C USA %D 2010 %F SisLab:3825 %P 435-442 %T Imagination-based Image Search System with Dynamic Query Creation and Its Application %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3825/ %0 Conference Paper %A Nguyen, Diep Thi Ngoc %A Sasaki, Shiori %A Kiyoki, Yasushi %B the First Indonesian-Japanese Conference on Knowledge Creation and Intelligent Computing (KCIC 2012) %C Surabaya, Indonesia %D 2012 %F SisLab:3822 %T Imagination-based Travel Designing System with 5D World PicMap %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3822/ %0 Journal Article %@ 2192-6611 %A Nguyen, Diep Thi-Ngoc %A Kiyoki, Yasushi %D 2017 %F SisLab:3786 %J International Journal of Multimedia Information Retrieval %N 2 %P 175-188 %T Multicontext-adaptive indexing and search for large-scale video navigation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3786/ %V 6 %X Many multimedia retrieval tasks are faced with increasingly large-scale datasets and variously changing preferences of users in each query. There are at least three distinctive contextual aspects comprised to form a set of preferences of a user at each query time: content, intention, and response time. A content preference refers to the low-level or semantic representations of the data that a user is interested in. An intention preference refers to how the content should be regarded as relevant. And a response time preference refers to the ability to control a reasonable wait time. This paper features the dynamic adaptability of a multimedia search system to the contexts of its users and proposes a multicontext-adaptive indexing and search system for video data. The main contribution is the integration of context-based query creation functions with high-performance search algorithms into a unified search system. The indexing method modifies inverted list data structure in order to construct disk-resident databases for large-scale data and efficiently enables a dynamic pruning search mechanism on those indices. We implement a frame-wise video navigation system as an application of the indexing and search system using the a 2.14 TB movie dataset. Experimental studies on this system show the effectiveness of the proposed pruning search method when dealing with dynamic contexts and its comparative high search performance. %0 Journal Article %A Nguyen, Dinh Chinh %A Luu, Manh Ha %A Guanghao, Sun %A Le, Quoc Anh %A Pham, Thi Viet Huong %A Tran, Anh Vu %A Tran, Trong Hieu %A Tran, Duc Tan %A Nguyen, Vu Trung %A Koichiro, Ishibashi %A Nguyen, Linh Trung %D 2020 %F SisLab:4246 %I ELSEVIER %J Biomedical Signal Processing and Control %T Short time cardio-vascular pulses estimation for dengue fever screening via continuous-wave Doppler radar using empirical mode decomposition and continuous wavelet transform %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4246/ %0 Journal Article %@ 1813-9663 %A Nguyen, Dinh Nghia %A Nguyen, Hoai Son %D 2018 %F SisLab:3091 %I Vietnam Academy of Science and Technology %J Journal of Computer Science and Cybernetics %N 2 %P 145-159 %T Congestion Control Algorithm for Message Routing in Structured Peer-to-Peer Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3091/ %V 34 %X In structured peer-to-peer (P2P) networks, every node must perform message routing to deliver query messages to destination nodes. Therefore, they often have to process a large number of query messages quickly and efficiently. When the number of query messages sending to a node exceeds its routing capacity, a local congestion on the node occurs and reduces the information pro- cessing of the other nodes in the network. This paper proposes a congestion control algorithm for message routing in structured P2P networks by changing the routing table of nodes in the network to route the packet to the destination without going through congested nodes. The performance of the proposed the method has been evaluated and compared with the conventional Chord protocol. The result shows that our proposed method improves the query success rate significantly while minimizing the number of packets generated during congestion control process. %0 Journal Article %@ 1813-9663 %A Nguyen, Dinh Viet %D 2002 %F SisLab:2227 %J Journal of Computer Science and Cybernetics %N 4 %P 339-346 %T Simulation-based Evaluation of Performance of a Snoop TCP Scheme over Networks with Wireless Links %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2227/ %X TCP is a transport protocol designed and tuned to work well in fixed networks, however TCP performance is degraded severely in a network with wireless links. To enhance TCP performance in such networks, some proposals have been made in the last several years. Among them Snoop TCP proposal is one of the most important ones. Our paper investigates the effect of various error rates of wireless link on the performance of snoop TCP; normal TCP is also put to our investigation to make a comparison. We use the network simulator ns-2, version 2.1b9 extensively. Our simulation results show that in comparison with normal TCP, snoop TCP enhances normalized throughput of LAN and WAN with wireless link 137% and 427% respectively. %0 Journal Article %@ 1813-9663 %A Nguyen, Dinh Viet %D 2002 %F SisLab:2228 %J Journal of Computer Science and Cybernetics %N 1 %P 87-96 %T So sánh hiệu suất của các kế hoạch kiểm soát lỗi trong mạng truyền thông máy tính %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2228/ %X This paper analyzes quantitatively link-by-link and end-to-end error control schemes in computer communication networks. Two measures used for performance evaluations are mean packet delay time and throughput of the networks. With two specific examples, performance comparisons are made for two cases: traditional computer networks (low-speed) and high-speed (gigabit/sec) computer networks. The results is that: link-by-link error control scheme is superior than end-to-end error control scheme for low-speed computer networks, by contrast, for high-speed computer networks end-to-end scheme is superior than link-by-link scheme. %0 Journal Article %@ 1813-9663 %A Nguyen, Dinh Viet %D 2002 %F SisLab:2229 %J Journal of Computer Science and Cybernetics %N 2 %P 187-200 %T TCP Enhancements and Performance over Networks with Wireless Links %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2229/ %X Transmission Control Protocol (TCP) uses end-to-end flow, congestion, and error control mechanisms to provide reliable delivery over the Internet. During many years of extensive use in inter-networks, all weak-points of TCP are discovered and tuned, so that it performs very well in traditional network (wired network). However, inter-networks growth explosively and may consist of networks with wireless links characterized by high and sporadic bit error rate and intermittent connectivity due to handoff. TCP performance is degraded severely in such networks. Many studies have been carrying out to improve the performance of TCP in networks with wireless links. Some solutions are proposed and implemented. This paper presents four main issues: Congestion avoidance and control mechanisms implemented in traditional TCP protocol; Step-by-step enhancements in traditional TCP; Characteristics of wireless links degrade TCP performance severely; and mechanisms to enhance TCP performance over network with wireless links. %0 Conference Paper %A Nguyen, Dinh Viet %B Hội nghị Khoa học Trường Đại học Khoa học Tự nhiên, ĐHQGHN %C Hà Nội, Việt Nam %D 1998 %F SisLab:2996 %P 23-30 %T Về hệ mã mật khoá công khai %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2996/ %X At present, most of information resources are on computer's memory, and computers communicate with each other via computer networks. Those communication channels are insecure in the sense that, when a sender A sends a message to receiver B, an enemy could receive the message physically. The most important role of a cryptosystem is to make the enemy unable to know the content of the message - to guarantee the Secrety and the Authenticity of information. The cryptosystems based on symmetric algorithms use the same key for both encryption and decryption process, so the key must be kept secret. A sender and a receiver have to agree about the secret key before communication or have to use a secure channel to send secret key from the sender to the receiver. It is a matter of fact that it is not always possible. The cryptosystems based on asymmetric algorithms use one key - KB for encryption and another key - kB for decryption. A Key generator generates a pair (KB, kB) basing on the initial conditions. When B wants to send A a message, he/she uses Key generator to generate a pair (KB, kB) then sends KB (it is called Public-Key) to A, so A can use it to encrypt the message. When receiving encrypted message, B uses the secret key kB to decrypt it. The asymmetric cryptosystems must guarantee that, if the enemy receives the encrypted message and the key KB associated with it, it is not practically possible for him to calculate the secret key kB to decrypt the message. In other word, that is an intractable problem for him. The name Public-Key comes from that sense. %0 Conference Paper %A Nguyen, Dinh Viet %A Le, Dinh Danh %B Hội thảo Quốc gia lần thứ XI: “Một số vấn đề chọn lọc của Công nghệ thông tin và Truyền thông” %C Hue, Viet Nam %D 2008 %F SisLab:2986 %P 239-248 %T ĐÁNH GIÁ HIỆU SUẤT CHIẾN LƯỢC QUẢN LÝ HÀNG ĐỢI A-RIO BẰNG BỘ MÔ PHỎNG MẠNG NS %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2986/ %X Để đảm bảo chất lượng dịch vụ (QoS) truyền thông đa phương tiện trên Internet, trên tất cả các nút mạng (router) mà các luồng thông tin đi qua cần áp dụng các chính sách quản lý hàng đợi và chính sách phục vụ phù hợp. Các chính sách này đảm bảo cho các luồng thông tin đa phương tiện có thể đạt được thông lượng đủ lớn, thời gian trễ trung bình và thăng giáng của thời gian trễ (jitter) của các gói tin đủ nhỏ. Ngoài ra cần đảm bảo sự phục vụ với các mức ưu tiên khác nhau cho các luồng thông tin đa phương tiện khác nhau. Người ta đã nghiên cứu và áp dụng vào thực tế nhiều chiến lược quản lý hàng đợi động (AQM) đáp ứng được các yêu cầu nói trên, đó là RED và các cải tiến của nó là A-RED, RIO và A-RIO... Báo cáo này trình bày các kết quả nghiên cứu của chúng tôi về A-RIO bằng bộ mô phỏng mạng NS. %0 Conference Paper %A Nguyen, Dinh Viet %A Le, Minh Khoi %A Le, Thi Thuy %A Do, Thi Bac %A Nguyen, The Cong %B Hội thảo Quốc gia lần thứ XV Một số vấn đề chọn lọc của Công nghệ thông tin và Truyền thông %C Việt Nam %D 2012 %F SisLab:2983 %P 52-59 %T Capacity Evaluation of Mobile Ad Hoc Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2983/ %X One of the main limitations of Ad hoc networks is limited capacity; This is mainly due to interference which limits the spatial reuse of wireless channels. It acts as a problem to the deployment of Ad hoc networks. Evaluation of the Capacity of Ad hoc networks is difficult because of the complexity of routing and MAC protocols, etc. Until now, many methods for Ad hoc network capacity evaluation have been developed. However, most of them have flaws, since some simplification assumptions must be made to make capacity evaluation problem tractable. This paper introduces some approximation algorithms to evaluate the capacity of Ad hoc Network, and presents our capacity evaluation results obtained by simulating ad hoc network in various scenarios with the use of NS-2 network simulator. %0 Conference Paper %A Nguyen, Dinh Viet %A Nguyen, Manh Ha %B Hội thảo Quốc gia lần thứ XII: “Một số vấn đề chọn lọc của Công nghệ thông tin và Truyền thông” %C Bien Hoa, Viet Nam %D 2010 %F SisLab:2985 %P 93-106 %T ĐÁNH GIÁ ẢNH HƯỞNG CỦA TRỌNG SỐ HÀNG ĐỢI ĐẾN HIỆU SUẤT CỦA CHIẾN LƯỢC QUẢN LÝ HÀNG ĐỢI A-RIO %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2985/ %X Để đảm bảo chất lượng dịch vụ (QoS) truyền thông đa phương tiện trên Internet, cần áp dụng các chính sách quản lý hàng đợi và chính sách phục vụ phù hợp tại tất cả các nút mạng (router) mà các luồng thông tin đi qua. Các chính sách này đảm bảo cho các luồng thông tin đa phương tiện có thể đạt được thông lượng đủ lớn, thời gian trễ trung bình và thăng giáng của thời gian trễ (jitter) của các gói tin đủ nhỏ. Ngoài ra cần đảm bảo sự phục vụ với các mức ưu tiên khác nhau cho các luồng thông tin đa phương tiện khác nhau. Nhiều chiến lược quản lý hàng đợi động (AQM) đã được nghiên cứu và áp dụng vào thực tế, đáp ứng được các yêu cầu nói trên, đó là RED và các cải tiến của nó là A-RED, RIO và A-RIO. Mỗi chiến lược nói trên đều phải dựa vào kích thước hàng đợi trung bình mà nó tính được để hoạt động một cách có hiệu quả nhất. Báo cáo này trình bày các kết quả nghiên cứu của chúng tôi về ảnh hưởng của trọng số hàng đợi đến hiệu suất của chiến lược quản lý hàng đợi A-RIO bằng bộ mô phỏng mạng NS-2. Đây là sự phát triển tiếp theo và sâu hơn các nghiên cứu đã được báo cáo tại Hội thảo Quốc gia về CNTT và Truyền thông, được tổ chức tại Huế, tháng 6/2008 và đã được đăng toàn văn trong Kỷ yếu của Hội thảo. %0 Conference Paper %A Nguyen, Dinh Viet %A Nguyen, Manh Ha %A Nguyen, Minh Nguyet %B Hội thảo Quốc gia lần thứ sáu: Một số vấn đề chọn lọc của Công nghệ Thông tin %C Thái Nguyên, Việt Nam %D 2005 %F SisLab:2993 %P 484-491 %T A comparison of caching solutions to improve web cache consistency in scalable network %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2993/ %X The break out of the Internet has made the World Wide Web become “world wide wait” due to dramatically increased load on the network infrastructure and on individual web servers. To ameliorate these burdens, web caching, a technology that hold any retrieved object can reduce network, server load and the latency of response substantially. However, there is a serious disadvantage that the object returned to client by cache maybe stale and inconsistence with the one on server. This paper discusses several caching algorithms and architectures to limit the degree of staleness. We also propose an integrated web cache solution providing fairly tight bounds on the staleness and evaluate this design based on simulation experiences compared to that of some other solutions. %0 Conference Paper %A Nguyen, Dinh Viet %A Phan, Huu Dung %A Doan, Cao Thanh %A Ho, Duc Ai %B Hội thảo Quốc gia lần thứ XIV: “Một số vấn đề chọn lọc của Công nghệ thông tin và Truyền thông %C Can Tho, Viet Nam %D 2012 %F SisLab:2984 %P 64-80 %T ĐÁNH GIÁ HIỆU NĂNG CỦA CÁC GIAO THỨC ĐỊNH TUYẾN TRONG MẠNG MANET %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2984/ %X MANET (Mobile wireless Ad hoc NETwork) là một loại mạng bao gồm các nút mạng có thể di động, chúng sử dụng đường truyền không dây để truyền thông với nhau một cách trực tiếp hoặc thông qua một số nút trung gian đóng vai trò chuyển tiếp gói tin như các router, mà không cần dựa trên cơ sở hạ tầng mạng có dây và các đơn vị quản trị trung tâm. Với MANET, vì các nút mạng đều có thể di chuyển tuỳ ý, nên không thể sử dụng các giao thức và thuật toán định tuyến vốn được sử dụng phổ biến trên mạng có dây. Thực hiện được việc định tuyến một cách có hiệu quả là bài toán quan trọng và nhiều thách thức nhất đối với những người nghiên cứu MANET. Trong bài báo này chúng tôi trình bày các kết quả nghiên cứu, đánh giá và so sánh hiệu suất của các giao thức định tuyến AODV, DSR, DSDV và OLSR theo mức độ linh động của các nút mạng; hai tham số hiệu suất được chúng tôi sử dụng là chi phí tìm đường và tỉ lệ phân phát gói tin thành công. Các thực nghiệm bằng bộ mô phỏng NS2 được thực hiện với hai mô hình chuyển động là Random Waypoint và Random Walk. Các kết quả thu được của chúng tôi cho thấy rõ hiệu quả hoạt động của AODV, DSR, DSDV và OLSR theo các mức độ linh động khác nhau của các nút mạng. Nhận xét của chúng tôi về các kết quả thu được cũng chính là khuyến nghị về việc áp dụng các giao thức định tuyến cụ thể. %0 Conference Paper %A Nguyen, Dinh Viet %A Vu, Duy Loi %B Hội thảo Khoa học Quốc gia lần thứ nhất về Nghiên cứu, Phát triển và Ứng dụng Công nghệ Thông tin và Truyền thông (ICT.rda) %C Hà Nội, Việt Nam %D 2003 %F SisLab:2995 %P 429-436 %T ĐÁNH GIÁ HIỆU SUẤT HOẠT ĐỘNG CỦA SNOOP TCP TRONG MẠNG KHÔNG ĐỒNG NHẤT BẰNG PHƯƠNG PHÁP MÔ PHỎNG %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2995/ %X Bài báo này trình bày các kết quả nghiên cứu của chúng tôi về đánh giá bằng mô phỏng hiệu suất của Snoop TCP trong mạng không đồng nhất, gồm các mạng có dây và không dây. Đầu tiên, chúng tôi tóm tắt một số kết quả nghiên cứu cải tiến hiệu suất TCP trong mạng có đường truyền không dây; sau đó, mô tả các chức năng cơ bản của Snoop TCP và cấu hình của mạng mô phỏng. Chúng tôi đã thực hiện nhiều thí nghiệm mô phỏng bằng bộ mô phỏng mạng NS-2 để nghiên cứu kỹ ảnh hưởng của độ dài trung bình trạng thái xấu của đường truyền không dây cũng như đơn vị thời gian của đồng hồ phát lại cục bộ (SnoopTick) đối với hiệu suất của Snoop TCP; để so sánh, chúng tôi cũng đã nghiên cứu TCP truyền thống. Các kết quả nghiên cứu bằng mô phỏng của chúng tôi cho thấy rằng: Snoop TCP tốt hơn TCP thông thường còn Snoop TCP sử dụng giá trị SnoopTick mà chúng tôi đề xuất đạt được hiệu suất cao hơn Snoop TCP đã được công bố là 150%. %0 Conference Paper %A Nguyen, Dinh Viet %A Vu, Duy Loi %A Vu, Duc Trung %B Hội thảo khoa học Quốc gia lần thứ II: Nghiên cứu cơ bản và Ứng dụng Công nghệ Thông tin" (FAIR) %C Thành phố Hồ Chí Minh, Việt Nam %D 2006 %F SisLab:2989 %P 62-73 %T ĐẢM BẢO CHẤT LƯỢNG DỊCH VỤ CHO TRUYỀN THÔNG ĐA PHƯƠNG TIỆN THEO KIẾN TRÚC DIFFSERV %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2989/ %X Khác với các ứng dụng truyền thống trên Internet thường là thuần văn bản, các ứng dụng truyền thông đa phương tiện truyền tải nhiều loại dữ liệu khác nhau như âm thanh, hình ảnh, văn bản,...Đặc điểm của các ứng dụng này là có thể chấp nhận sự mất mát thông tin trong một giới hạn nhất định, nhưng đòi hỏi tính thời gian thực (real-time), nghĩa là yêu cầu độ trễ (delay) và thăng giáng độ trễ (jitter) nhỏ. Việc đảm bảo chất lượng dịch vụ (QoS) là vấn đề hết sức khó khăn và mang tính quyết định đến sự tồn tại, phát triển của các ứng dụng truyền thông đa phương tiện, vì vậy rất nhiều nghiên cứu đã được tiến hành nhằm tìm ra các giải pháp hữu hiệu để đảm bảo QoS như IntServ, Diffserv.... Trong báo cáo này, chúng tôi sẽ trình bày các vấn đề sau: (1) Các ứng dụng truyền thông đa phương tiện, (2) Yêu cầu QoS đối với truyền thông đa phương tiện, (3) Các phương pháp đảm bảo QoS IntServ (Integrated Services) và DiffServ (Differentiated Services) trong đó chủ yếu trình bày về DiffServ, và (4) các kết quả nghiên cứu của chúng tôi, sử dụng bộ mô phỏng mạng NS để đánh giá hiệu quả của phương pháp DiffServ đến QoS của truyền thông đa phương tiện. %0 Conference Paper %A Nguyen, Do Van %A Vuong, Tung Long %A Kieu, Hai Dang %A Pham, Linh %A Le, Thanh Ha %B IEEE International Conference on Systems, Man, and Cybernetic (IEEE SMC2018) %D 2018 %F SisLab:3294 %T Vision Memory for Target Object Navigation using Deep Reinforcement Learning: An Emperical Study %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3294/ %0 Conference Paper %A Nguyen, Dong Dong %A Bui, Ngoc Dung %A Hoang, Xuan Tung %A Tran, Nguyen Cac %B INTERNATIONAL CONFERENCE ON APPLICATIONS OF ARTIFICIAL INTELLIGENCE IN TRANSPORTATION %D 2020 %F SisLab:4277 %P 33-40 %T POSE INVARIANT FACE RECOGNITION SYSTEM %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4277/ %X Face recognition has become popular in security surveillance systems. These systems work well only when the face is in frontal and the system itself has considerable training data. This paper proposes a method of face recognition with various view using Generative Adversarial Networks (GAN). In this method, a pre-trained model of face is used to generate the training data of different pose variations from several frontal faces using GAN. These faces will be used as input of the FaceNet and Support Vector Machine to be trained to perform the task of face recognition. The experiment results demonstrate the accuracy of the proposed method in the case of lacking the training data. This paper focus on the applicability of Generative Adversarial Network in Face Recognition System by applying the GAN-based method in Face Recognition System to increase the number of training samples and detect the faces of different poses that was not possible with the conventional Face Recognition System. The comparison between Face Recognition System using GAN and conventional Face Recognition System was presented. %0 Conference Paper %A Nguyen, Duc Anh %A Do, Minh Kha %A Pham, Thi To Nga %A Pham, Ngoc Hung %B 2021 IEEE International Conference on Research, Innovation and Vision for the Future (RIVF 2021) %C Hanoi %D 2021 %F SisLab:4468 %T An Autoencoder-based Method for Targeted Attack on Deep Neural Network Models %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4468/ %0 Conference Paper %A Nguyen, Duc Anh %A Pham, Ngoc Hung %B The Eighth International Symposium on Information and Communication Technology (SoICT 2017) %C Nha Trang %D 2017 %F SisLab:2636 %T A Test Data Generation Method for C/C++ Projects %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2636/ %X This research proposes an automated test data generation method for C/C++ projects to generate the lower number of test data while gaining higher code coverage in comparison with KLEE, CAUT, PathCrawler, and CREST. In order to do that, the proposed method contributes an algorithm named loop depth frst search by combining both static testing and concolic testing together. Besides, the paper also provides an improvement symbolic execution for avoiding the initial test data problem in the concolic testing. Currently, a tool supporting the proposed method has been developed and applied to test on different C/C++ projects in several software companies. The experimental results show the higher coverage with the lower number of test data compared with the existing methods. The experimental results display the effectiveness and practical usefulness of the proposed method for automated test data generation in practice %0 Conference Paper %A Nguyen, Duc Anh %A Pham, Ngoc Hung %A Nguyen, Viet Ha %B NICS: 3rd National Foundation for Science and Technology Development Conference on Information and Computer Science %C Da Nang city, Viet Nam %D 2016 %F SisLab:2002 %T A Method for Automated Unit Testing of C Programs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2002/ %X This research proposes an automated test case generation method for C functions. In this method, the source code is transformed into a control flow graph corresponding to the given coverage criterion. After that, a list of feasible test paths is discovered by traversing the control flow graph using backtracking algorithm, symbolic execution, and Z3 solver. We also generate test cases for functions containing one loop or two-nested loop. A tool supporting the proposed method has been developed and applied to test on some C functions. The experimental results show the high coverage with the minimum number of test cases, the ability to improve the total time of the test case generation with a specified coverage criterion, and the increasing precision of checking the feasibility of test paths if comparing with the random technique. The experimental results display the potential usefulness of this tool for automated test case generation in practice. %0 Journal Article %@ 10.1142/S0218194019500402 %A Nguyen, Duc Anh %A Pham, Ngoc Hung %A Vo, Dinh Hieu %A Tran, Nguyen Huong %D 2019 %F SisLab:3734 %I World Scientic Publishing Company %J International Journal of Software Engineering and Knowledge Engineering %N 9 %P 1279-1312 %T Improvements of Directed Automated Random Testingin Test Data Generation for C++ Projects %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3734/ %V 29 %X This paper improves the breadth-¯rst search strategy in directed automated random testing (DART) to generate a fewer number of test data while gaining higher branch coverage, namely Static DART or SDART for short. In addition, the paper extends the test data compilation mechanism in DART, which currently only supports the projects written in C, to generate test data for C++ projects. The main idea of SDART is when it is less likely to increase code coverage with the current path selection strategies, the static test data generation will be applied with the expectation that more branches are covered earlier. Furthermore, in order to extend the test data compilation of DART for C++ context, the paper suggests a general test driver technique for C++ which supports various types of parameters including basic types, arrays, pointers, and derived types. Currently, an experimental tool has been implemented based on the proposal in order to demonstrate its e±cacy in practice. The results have shown that SDART achieves higher branch coverage with a fewer number of test data in comparison with that of DART in practice. %0 Journal Article %@ 0218-1940 %A Nguyen, Duc Anh %A Tran, Nguyen Huong %A Vo, Dinh Hieu %A Pham, Ngoc Hung %D 2019 %F SisLab:3502 %I IJSEKE %J International journal of software engineering and knowledge engineering (IJSEKE) %T ON IMPROVEMENTS OF DIRECTED AUTOMATED RANDOM TESTING IN TEST DATA GENERATION FOR C++ PROJECTS %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3502/ %X This paper improves the breadth-first search strategy in directed automated random testing (DART) to generate a fewer number of test data while gaining higher branch coverage, namely Static DART or SDART for short. In addition, the paper extends the test data compilation mechanism in DART, which currently only supports the projects written in C, to generate test data for C++ projects. The main idea of SDART is when it is less likely to increase code coverage with the current path selection strategies, the static test data generation will be applied with the expectation that more branches are covered earlier. Furthermore, in order to extend the test data compilation of DART for C++ context, the paper suggests a general test driver technique for C++ which supports various types of parameters including basic types, arrays, pointers, and derived types. Currently, an experimental tool has been implemented based on the proposal in order to demonstrate its efficacy in practice. The results have shown that SDART achieves higher branch coverage with a fewer number of test data in comparison with that of DART in practice. %0 Journal Article %A Nguyen, Duc Bach %A Chu, Duc Ha %A Vu, Le Dieu Huong %A Phi, Thi Cam Mien %F SisLab:4478 %J Tạp chí Khoa học Nông nghiệp Việt Nam %T ĐỊNH DANH VÀ XÁC ĐỊNH ĐẶC ĐIỂM SINH TRƯỞNG CỦA CHỦNG VI KHUẨN LAM Arthrospira platensis PHÂN LẬP TỪ HỒ VĂN QUÁN %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4478/ %0 Conference Paper %A Nguyen, Duc Canh %A Dang, Cao Cuong %A Le, Sy Vinh %A Bui, Quang Minh %A Hoang, Thi Diep %B 2020 12th International Conference on Knowledge and Systems Engineering (KSE) %C Can Tho Viet Nam %D 2020 %F SisLab:4215 %P 324-329 %T pQMaker: empirically estimating amino acid substitution models in a parallel environment %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4215/ %X Amino acid substitution matrices are central to the model-based methods for reconstructing evolutionary trees from amino acid sequences. QMaker is an efficient method for estimating general time-reversible amino acid substitution matrices from a large biological dataset containing thousands of protein alignments using maximum likelihood principle. It allows researchers to build an amino acid substitution model on their own to best fit their subsequent phylogenetic analyses. In this work, we propose an approach to parallelize computation in QMaker, named pQMaker. Moreover, we provide an open-source message passing interface implementation for pQMaker (https://github.com/canhnd58/IQ-TREE/tree/pqmaker) built upon the latest IQ-TREE package. Experiments on benchmark data sets show that our implementation has significant speed gains compared with the original QMaker. %0 Conference Paper %A Nguyen, Duc Linh %A Man, Duc Chuc %A Bui, Quang Hung %A Nguyen, Thi Nhat Thanh %B The 8th International Conference on Knowledge and Systems Engineering (KSE) %C Hanoi, Vietnam %D 2016 %F SisLab:2776 %K Atmospheric measurements;Correlation;Data models;Filling;Monitoring;Pollution measurement;Training;PM10;abnormal detection;environmental data;missing filling %P 321-326 %T Standardization procedure for automatic environmental data: A case study in Hanoi, Vietnam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2776/ %X In Vietnam, environmental data collected from ground-based stations may contain abnormal or missing values due to several problems during operation, i.e. sensor's problems. This paper proposes a standardization procedure which try to detect unusual values and fill in missing data. Experiments were conducted for PM10 data. Two datasets measured in 01/2011 and 01/2012 at Nguyen Van Cu station in Hanoi, Vietnam is used for experiments. For the abnormal detection process, unusual data can be informed to the data analyzers at ground stations for judging. For the missing filling process, the first dataset is used as training dataset to construct regression models for predicting missing data, the second dataset is used as testing data. In the worst case, suppose 100% PM10 is missing, Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) are 51 µg/m3 and 45% respectively. Correlation coefficient (R) between original PM10 data and predicted PM10 data is 0.56. In addition, different scenarios taking account of percentage of missing data of the whole testing dataset are also considered. Experimental results showed that it is best to perform missing filling process on datasets that contain 10% to 30% of missing data. For this case, RMSE ranges from 15–25 µg/m3 and MAPE varies from 5 to 13%. %0 Journal Article %@ 2588-1086 %A Nguyen, Duc Thao %A Nguyen, Viet Anh %A Le, Thanh Ha %A Ngo, Thi Duyen %D 2021 %F SisLab:4480 %I VNU %J VNU Journal of Science: Computer Science and Communication Engineering %N 1 %T Robustify Hand Tracking by Fusing Generative and Discriminative Methods %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4480/ %V 37 %X With the development of virtual reality (VR) technology and its applications in many fields, creating simulated hands in the virtual environment is an e ective way to replace the controller as well as to enhance user experience in interactive processes. Therefore, hand tracking problem is gaining a lot of research attention, making an important contribution in recognizing hand postures as well as tracking hand motions for VR’s input or human machine interaction applications. In order to create a markerless real-time hand tracking system suitable for natural human machine interaction, we propose a new method that combines generative and discriminative methods to solve the hand tracking problem using a single RGBD camera. Our system removes the requirement of the user having to wear to color wrist band and robustifies the hand localization even in di cult tracking scenarios. %0 Conference Paper %A Nguyen, Duc Trung %A Nguyen, Diep Thi Ngoc %A Kiyoki, Yasushi %B 5th Workshop on Multimedia for Cooking and Eating Activities (ACMM CEA2013), ACM 2013 %C Barcelona, Spain %D 2013 %F SisLab:3821 %P 15-20 %T A Regional Food's Features Extraction Algorithm and Its Application %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3821/ %X Automatically detecting food's taste is a non-trivial part. However, we realize that the taste of food can be extracted by directly analyzing recipes by the ingredients and the amount of them in the recipes. In this paper, we present a food analysis system to discover the taste of foods and to better understand the featured ingredients in each specific geographical region. The main features of this system are (1) to extract dominant ingredients and tastes in a region by analyzing the ingredients' frequency and its uniqueness, and (2) to transform user's existing materials or original recipe to a new recipe according to a targeted taste. To examine the feasibility and applicability of the algorithm, we have developed a web-based application with a recipe database collected from approximately 200 recipes in over 7 regions of Japan: Hokkaido-Tohoku, Kanto, Kansai, Shikoku, Tyubu, Kyusyu-Okinawa and Tyugoku. %0 Conference Paper %A Nguyen, Duc-Hieu %A Le, Van-Vinh %A Nguyen, Thi-Hanh %A Dang, Duc Hanh %B ICSSE: the 2021 International Conference on System Science and Engineering %D 2021 %F SisLab:4665 %P 360-366 %T A Method to Ensure Compliance with Attribute and Role Based Access Control Policy for Executing BPMN Models %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4665/ %0 Conference Paper %A Nguyen, Duy Khuong %A Than, Khoat %A Ho, Tu Bao %B 2013: IEEE RIVF International Conference on Computing and Communication Technologies (RIVF2013) %C Hanoi, VIETNAM %D 2013 %F SisLab:756 %P 47-52 %T Siplicial Nonnegative Matrix Factorization %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/756/ %0 Journal Article %@ 2234-4624 %A Nguyen, Duy Tan %A Longzhe, Han %A Nguyen, Dinh Viet %A Minho, Jo %D 2012 %F SisLab:2225 %J Smart Computing Review %N 5 %P 360-369 %T An Improved LEACH Routing Protocol for Energy-Efficiency of Wireless Sensor Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2225/ %V 2 %X A wireless sensor network (WSN) is a wireless network consisting of a base station (BS) and multiple micro-sensor nodes. The micro-sensor nodes run on batteries that are not recharged during the network’s lifetime. Therefore, prolonging WSN lifetime is a significant problem. There are some energy-efficient hierarchical routing protocols designed especially for WSNs, such as sensor protocols for information negotiation (SPIN), low-energy adaptive clustering hierarchy (LEACH) and LEACH-centralized (LEACH-C). In this article, we propose an improvement on the LEACH routing protocol to decrease energy consumption and prolong network lifetime. LEACH-Distance Energy (LEACH-DE) selects a cluster head node not only by considering residual energy of the node greater than the average residual energy level of nodes in network, but also by examining the geometric distance between the candidate node to the BS as a key parameter. The results of simulation show our proposed algorithm outperforms LEACH and LEACH-C in terms of conserving energy and prolonging WSN lifetime. %0 Conference Paper %A Nguyen, Duy Tan %A Nguyen, Dinh Viet %B Kỷ yếu Hội nghị Khoa học Công nghệ Quốc gia lần thứ VIII về Nghiên cứu cơ bản và ứng dụng Công nghệ thông tin (FAIR'8) %D 2015 %F SisLab:2232 %P 61-71 %T CAEER: Channel Assignment and Energy-Efficient Routing Protocol in Cognitive Radio Adhoc Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2232/ %V 2 %X One of the main challenging factors for designing routing protocols in cognitive radio ad-hoc network (CRAHNs) is how to select route for which can save energy battery for devices and avoid interference to primary users (PU). To solve this problem. In this paper, we propose channel assignment and energy-efficient routing (CAEER) protocol in multi-hop and multi-channel CRAHNs. The proposed routing scheme suggests an intelligent multi-channel selection function, which is used to assign an appropriate channel for link to build route based on interference avoidance to PUs. Moreover, our routing metrics also consider necessarily the path which reduces energy consumption and control overhead. Our simulation results show that CAEER performs better than weighted cumulative expected transmission time (WCETT) and Multi-radio Multichannel Ad-Hoc on-Demand Distance Vector (MM-AODV) protocols in term of energy consumption and the control overhead. %0 Conference Paper %A Nguyen, Duy Tan %A Nguyen, Dinh Viet %B 2014: Hội nghị quốc gia lần thứ VII Nghiên cứu cơ bản và ứng dụng Công nghệ thông tin (FAIR 2014) %C Thai Nguyen, Vietnam %D 2014 %F SisLab:1314 %T DFCB: Data Fusion and Chain-Based Routing Protocol for Energy-Efficient in Wireless Sensor Network %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1314/ %0 Conference Paper %A Nguyen, Duy Tan %A Nguyen, Dinh Viet %B 2014: The 6th International Conference on Knowledge and Systems Engineering (KSE 2014) %C Hanoi, Vietnam %D 2014 %F SisLab:438 %P 61-77 %T DFTBC: Data Fusion and Tree-Based Clustering Routing Protocol for Energy-Efficient in Wireless Sensor Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/438/ %0 Conference Paper %A Nguyen, Duy Tan %A Nguyen, Dinh Viet %B ATC: the 2015 International Conference Advanced Technologies for Communications %C Ho Chi Minh city, Vietnam %D 2015 %F SisLab:1614 %P 314-319 %T SCBC: Sector-Chain Based Clustering Routing Protocol for Energy Efficiency in Heterogeneous Wireless Sensor Network %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1614/ %X One of the most important challenges in heterogeneous wireless sensor network (WSN) is to design a routing protocol that use energy efficiently to prolong the lifetime of the entire network due to the limited battery power of sensor nodes. To solve this problem, in this paper, we propose Sector-Chain Based Clustering (SCBC) Routing Protocol, in which, entire network area is divided into sectors (clusters), which balance to the number of nodes. SCBC further reduces energy dissipation of communication in network by constructing chain for each cluster with the chain leader as the cluster head (CH) or secondary cluster head (SCH) that has high residual energy and the shortest distance between candidate nodes and base station (BS) in network. In addition, SCBC improves energy efficiency by calculating the time length of each round in the steady data transmission phase for CHs or SCHs but still guarantee that it is alive in next round. Our simulation results show that the network lifetime of our proposed protocol can be improved about 70% and 20% in comparison with PEGASIS (Power-Efficient Gathering in Sensor Information System) and IEEPB (An Improved Energy-Efficient PEGASIS-Based) protocols, respectively. %Z IEEE ATC 2015 Conference Proceedings %0 Conference Paper %A Nguyen, Duy Tan %A Nguyen, Dinh Viet %B 2015: The 11th IEEE-RIVF International Conference on Computing and Communication Technologies (RIVF 2015) %C Can Tho, Vietnam %D 2014 %F SisLab:1311 %T SSTBC: Sleep Scheduled and Tree-Based Clustering Routing Protocol for Energy-Efficient in Wireless Sensor Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1311/ %0 Conference Paper %A Nguyen, Ha Anh Tuan %A Le, Sy Vinh %A Le, Si Quang %B 2012 Fourth International Conference on Knowledge and Systems Engineering (KSE) %C Danang, Vietnam %D 2012 %F SisLab:106 %K Clustering algorithms;Educational institutions;Genetics;Maximum likelihood detection;Probability;Process control;Quality control;genetics;genomics;maximum likelihood detection;quality control;sampling methods;Illumina BeadChips data;bad SNP sites;bad samples detection;computational methods;genetic variation;genotype caller;genotype data;human health;manual quality control;maximum likelihood method;microarray technologies;quality control process;raw genotype data generation;SNP;bad samples;genotype;quality control; %P 26 -33 %T A Maximum Likelihood Method for Detecting Bad Samples from Illumina BeadChips Data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/106/ %X Genotype data provide crucial information to understand effects of genetic variation to human health. Current microarray technologies are able to generate raw genotype data from thousands of samples across million of SNP sites. These raw data are processed by computational methods, called genotype caller, to obtain genotypes. Genotype calls of different callers might not be consistent due to noise of bad samples or SNPs. This requires a manual quality control step conducted by experts to remove bad samples or bad SNP sites. In this paper, we propose a maximum likelihood method to detect bad samples to improve the reliability of the results. Experiments with real data demonstrate the usefulness of our method in the quality control process. Thus, our method has the ability to reduce the number of samples that are requested to manually check by experts. %0 Journal Article %@ 1943-3581 %A Nguyen, Ha Huy Cuong %A Vijender Kumar, Solanki %A Doan, Van Thang %A Nguyen, Thanh Thuy %D 2017 %F SisLab:2909 %I DOI: 10.5296/npa.v9i1-2 %J International Journal Network Protocols and Algorithms %N 1-2 %P 71-84 %T Resource Allocation for Heterogeneous Cloud Computing %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2909/ %V 9 %0 Journal Article %@ 1344-8994 %A Nguyen, Ha Nam %A Choi, Gyoo-Seok %D 2015 %F SisLab:1472 %I International Information Institute %J Information Journal %N 7 %P 3245-3252 %T Smartphones Behavior Analysis based on Sequential Pattern Approaches %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1472/ %V 18 %0 Conference Paper %A Nguyen, Ha Thanh %A Nguyen, Viet Ha %A Vu, Viet Anh %B The 9th International Conference on Knowledge and Systems Engineering (KSE) %C Hue, Vietnam %D 2017 %F SisLab:2703 %P 30-35 %T A Knowledge Representation for Vietnamese Legal Document System %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2703/ %X Legal documents play an important role in the legal systems of every nation. In order to process information automatically and make use of knowledge from legal documents, we need an appropriate knowledge representation for them. In this research, we investigate properties of the Vietnamese legal document system and propose a representation for the documents and their interrelationships. The main challenges for this task are their unstability, limited validity period and complicated interrelationships. Properties and relationships which we design to represent in our knowlege base are selective in order to make use of information within legal documents. In order to represent our knowlege base, we compare three methods: Semantic Network, Production Rules, Frame Language and make use of their advantages in our representation. Moreover, we implement a service-oriented system providing knowledge querying service for other application systems based on our research result. %0 Journal Article %A Nguyen, Hai Chau %D 2017 %F SisLab:2532 %I Springer %J 9th International Conference on Computational Collective Intelligence %T Enhancing Cholera Outbreaks Prediction Performance in Hanoi, Vietnam Using Solar Terms and Resampling Data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2532/ %X A solar term is an ancient Chinese concept to indicate a point of season change in lunisolar calendars. Solar terms are currently in use in China and nearby countries including Vietnam. In this paper we propose a new solution to increase performance of cholera outbreaks prediction in Hanoi, Vietnam. The new solution is a combination of solar terms, training data resampling and classification methods. Experimental results show that using solar terms in combination with ROSE resampling and random forests method delivers high area under the Receiver Operating Characteristic curve (AUC), balanced sensitivity and specificity. Without interaction effects the solar terms help increasing mean of AUC by 12.66%. The most important predictor in the solution is Sun’s ecliptical longitude corresponding to solar terms. Among the solar terms, "frost descent" and "start of summer" are the most important. %0 Journal Article %@ 2475-1839 %A Nguyen, Hai Chau %D 2019 %F SisLab:3613 %I Taylor & Francis %J Journal of Information and Telecommunication %N 4 %P 494-507 %T Estimation of Air Temperature Using Smartphones in Different Contexts %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3613/ %V 3 %X Measuring air temperature at a high spatial resolution is very important for many applications including detection of urban heat islands. However, air temperature is currently measured by weather stations those are very sparse spatially. In this paper, we propose a new approach to estimate air temperature using smartphones in different contexts. Most of the smartphones are not equipped with air temperature sensors but they are all equipped with battery temperature sensors. When a smartphone is in idle state, its battery temperature is stable and correlated with ambient air temperature. Furthermore, it is often carried close to human body, e.g. in pockets of coats, trousers and in hand. Therefore we developed a new approach of using two linear regression models to estimate air temperature from the idle smartphones battery temperature given their in-pocket or out-of-pocket positions. Lab test results show that the new approach is better than an existing one in mean absolute error and coefficient of determination metrics. Advantages of the new approach include the simplicity of implementation on smartphones and the ability for creating maps of temperature distribution. However, this approach needs field tests on more smartphone models to achieve its robustness. %0 Conference Paper %A Nguyen, Hai Chau %B 10th Asian Conference on Intelligent Information and Database Systems (ACIIDS) %C Dong Hoi, Quang Binh %D 2018 %F SisLab:3217 %P 633-641 %T A New Approach to Estimate Urban Air Temperature Using Smartphones %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3217/ %X Measuring urban air temperature at a high spatial resolution is very important for many applications including detection of urban heat islands. However air temperature is currently measured by professional weather stations those are very sparse at 50 km spatial resolution or more. In this paper, we propose a new approach to estimate air temperature from smartphones. Most of the smartphones are not equipped with air temperature sensors. However they are all equipped with battery temperature sensors. When a smartphone is in idle state, its battery temperature is stable and correlated with air temperature around the smartphone. Therefore we have developed a linear regression model to estimate air temperature from the idle smartphones battery temperature. Experiment results show that the new approach is statistically comparable to an existing one using mean error, mean absolute error and coefficient of determination metrics. Advantages of the new approach include simplicity of implementation on smartphones and ability of creating maps of temperature distribution. %0 Journal Article %@ 1860-949X %A Nguyen, Hai Chau %D 2013 %F SisLab:189 %J Advanced Computational Methods for Knowledge Engineering Studies in Computational Intelligence %P 209-224 %T Parallelization of the Fast Multipole Method for Molecular Dynamics Simulations on Multicore Computers %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/189/ %V 479 %0 Conference Paper %A Nguyen, Hai Chau %A Kubo, Masatoshi %A Le, Viet Hai %A Yamamoto, Tomoyuki %B 13th Asian Conference on Intelligent Information and Database Systems 2021 (ACIIDS 2021) %C Phuket, Thailand %D 2021 %F SisLab:4389 %T Phase Prediction of Multi-principal Element Alloys Using Support Vector Machine and Bayesian Optimization %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4389/ %X Designing new materials with desired properties is a complex and time-consuming process. One of the challenging factors of the design process is the huge search space of possible materials. Machine learning methods such as k-nearest neighbours, support vector machine (SVM) and artificial neural network (ANN) can contribute to this process by predicting materials properties accurately. Properties of multi-principal element alloys (MPEAs) highly depend on alloys’ phase. Thus, accurate prediction of the alloy’s phase is important to narrow down the search space. In this paper, we propose a solution of employing support vector machine method with hyperparameters tuning and the use of weights values for prediction of the alloy’s phase. Using the dataset consisting of the experimental results of 118 MPEAs, our solution achieves the cross-validation accuracy of 90.2%. We confirm the superiority of this score over the performance of ANN statistically. On the other dataset containing 401 MPEAs, our SVM model is comparable to ANN and exhibits 70.6% cross-validation accuracy. %Z Title and abstract of the paper will probably changed upon reviewers' recommendation. Latest version will be updated soon. %0 Journal Article %@ 2196-8888 %A Nguyen, Hai Chau %A Kubo, Masatoshi %A Le, Viet Hai %A Yamamoto, Tomoyuki %D 2022 %F SisLab:4742 %I World Scientific %J Vietnam Journal of Computer Science %T Support Vector Machine-based Phase Prediction of Multi-principal Element Alloys %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4742/ %X Designing new materials with desired properties is a complex and time-consuming process. One of the most challenging factors of the design process is the huge search space of possible materials. Machine learning methods such as k-nearest neighbours, support vector machine (SVM) and artificial neural network (ANN) can contribute to this process by predicting materials properties accurately. Properties of multi-principal element alloys (MPEAs) highly depend on alloys’ phase. Thus, accurate prediction of the alloy’s phase is important to narrow down the search space. In this paper, we propose a solution of employing support vector machine method with hyperparameters tuning and the use of weighted values for prediction of the alloy’s phase. Using the dataset consisting of the experimental results of 118 MPEAs, our solution achieves a cross-validation accuracy of 90.2%. We confirm the superiority of this score over the performance of ANN statistically. On the other dataset containing 401 MPEAs, our SVM model is comparable to ANN and exhibits 70.6% cross-validation accuracy. We also found that additional variables, including average melting temperature and standard deviation of melting temperature, increase prediction accuracy by 3.34% in the best case. %0 Journal Article %@ 2194-5357 %A Nguyen, Hai Chau %A Le, Thi Ngoc Anh %D 2016 %F SisLab:1838 %I Springer %J Advanced Computational Methods for Knowledge Engineering Volume 453 of the series Advances in Intelligent Systems and Computing %P 195-212 %T Using Local Weather and Geographical Information to Predict Cholera Outbreaks in Hanoi, Vietnam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1838/ %V 453 %X In 2007, repeated outbreaks of cholera in Hanoi have raised the need to have up-to-date evidence on the impact of factors on cholera epidemic, which is essential for developing an early warning system. We have successfully built models to predict cholera outbreaks in Hanoi from 2001 to 2012 using Random Forests method. We found that geographical factors - the number of cholera cases of a district of interest and its neighbours - are very important to predict accurately cholera cases besides the weather factors. Among weather factors, temperature and relative humidity are the most important. We also found that prediction accuracy of our models, measured in adjusted coefficient of determination, will decrease by 0.0076 if prediction length increases by one day. %0 Conference Paper %A Nguyen, Hanh Phuc %A Truong, Ninh Thuan %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1531 %T Mô hình hóa và kiểm chứng các hệ thống thứ lỗi và cân bằng tải %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1531/ %X Trong thiết kế nói chung, tùy thuộc vào mức độ quan trọng, tính chất sống còn của hệ thống đòi hỏi người thiết kế phải đảm bảo tính liên tục trong hoạt động của hệ thống. Ví dụ như hệ thống điều khiển bay, hệ thống cơ sở dữ liệu của các ngân hàng,…Hệ thống thứ lỗi (fault tolerance systems) là khả năng đảm bảo cho hệ thống tiếp tục thực hiện các chức năng của mình khi một phần của hệ thống gặp sự cố. Nói cách khác, hệ thống có thể cung cấp các dịch vụ trong khi vẫn tồn tại lỗi. Nghiên cứu trình bày một phương pháp để mô hình hóa và kiểm chứng tính chất của hệ thống chịu lỗi và cân bằng tải. Bên cạnh đó việc xây dựng và thử nghiệm công cụ kiểm chứng tính chất của hệ thống chịu lỗi và cân bằng tải cũng sẽ được giới thiệu. %0 Conference Paper %A Nguyen, Hanh Phuc %A Truong, Ninh Thuan %A Le, Hong Anh %B 7th EAI International Conference on Context-Aware Systems and Applications (ICCASA 2018) %C Vietnam %D 2018 %F SisLab:3348 %T jFAT: An automation framework for web application testing %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3348/ %X Web technologies have developed rapidly because web application is currently leading the trends of software development. A webbased application is a program that is accessed over a network connection, rather than existing within a device’s memory, hence detecting its failures is different from other software systems. Many approaches and tools have been proposed for web testing, however, introducing new frameworks is still an emerging topic in this field. This paper proposes an automation framework running in Java platform for web testing, called jFAT, which integrates with Selenium and TestNG. The paper also illustrates the use of framework with the Bank application case study. %0 Book Section %A Nguyen, Hoai Son %B Applying Integration Techniques and Methods in Distributed Systems and Technologies %D 2019 %F SisLab:3448 %I IGI global %P 134-155 %T Topology Optimization for Heterogeneous DHT- Based Multicast %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3448/ %X Since the deployment of IP multicast remains restricted due to many practical and political issues, researchers have shifted focus to exploiting application-layer multicast for multicast data delivery. Recently there has been considerable interest in applying DHT routing algorithms to application-level multicast. However, early DHT-based multicast protocols are insufficient in addressing a number of technical issues such as heterogeneous capacity of nodes or node churn. In this chapter, the author describes a solution called BAM-Chord (i.e., Bandwidth Adaptive Multicast over Chord) that optimizes the topology of a multicast tree based on node bandwidth. In the proposed solution, node position (i.e., node identifier) on a BAM-Chord ring will be decided based on node bandwidth capacity such that it can build a wide and balanced multicast tree rooted at the source node. As a result, BAM-Chord protocol can utilize network resources of every node to reduce the depth of the multicast tree and take advantages of DHTs in maintaining the multicast tree. %0 Conference Paper %A Nguyen, Hoai Son %A Do, Xuan Anh %A Le, Hoang %A Nguyen, Van Hoang %A Duong, Quang Khai %A Nguyen, Xuan Viet Cuong %A Ngo, Minh Hoang %B The 2020 RIVF International Conference on Computing and Communication Technologies %C Ho Chi Minh city, Vietnam %D 2020 %F SisLab:3976 %T ECHONET Lite-based IoT Platform for Smart Homes %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3976/ %X In this paper, we introduce an ECHONET Lite based IoT platform which allows households to set up smart home environment with ease as well as allows service providers to deploy their IoT applications easily and securely. Our proposed platform is fully implemented with three main features: a simple process for integrating new IoT devices into smart home environment, a seamless data transfer mechanism between IoT devices and service providers, and a RESTful API for verifying and supporting service providers to deploy their IoT applications into our platform. The experimental results verified the practicability of our proposed IoT platform. %0 Book Section %A Nguyen, Hoai Son %A Makino, Yoshiki %A Lim, Azman Osman %A Tan, Yasuo %A Shinoda, Yoichi %B Lecture Notes in Computer Science %D 2013 %F SisLab:778 %I Springer Science \mathplus Business Media %P 159-166 %T Building High-Accuracy Thermal Simulation for Evaluation of Thermal Comfort in Real Houses %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/778/ %V 7910 %0 Conference Paper %A Nguyen, Hoai Son %A Makino, Yoshiki %A Lim, Yuto %A Tan, Yasuo %B 2017 4th NAFOSTED Conference on Information and Computer Science (NICS 2017) %C Hanoi, Vietnam %D 2017 %F SisLab:2678 %T Short-term prediction of energy consumption of air conditioners based on weather forecast %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2678/ %X In residential houses, air conditioners consume a lot of electrical energy. In order to improve energy efficiency for residential houses, short-term prediction of energy consumption of air conditioners is required. In this paper, we propose the use of our thermal simulation to simulate the change of room temperature based on weather forecast information and predict the energy consumption of an air conditioner in a residential house. In order to calculate solar radiation heat flux, which contributes a lot to the change of room temperature, we utilize a neural network model to predict global solar radiation using training data obtained from weather stations. We also utilize a PID control model to simulate the operation of air conditioners. The accuracy of our simulation is verified by experiments carried out at a real testbed house. %0 Journal Article %@ 0916-8516 %A Nguyen, Hoai Son %A Nguyen, Dinh Nghia %A Sugawara, Shinji %D 2019 %F SisLab:3447 %J IEICE Transactions on Communications %N 3 %P 545-556 %T A Dynamic-Clustering Backup Scheme for High-Availability Distributed File Sharing Systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3447/ %V E102-B %X DHT routing algorithms can provide efficient mechanisms for resource placement and lookup for distributed file sharing systems. However, we must still deal with irregular and frequent join/leave of nodes and the problem of load unbalancing between nodes in DHT-based file sharing systems. This paper presents an efficient file backup scheme based on dynamic DHT key space clustering in order to guarantee data availability and support load balancing. The main idea of our method is to dynamically divide the DHT network into a number of clusters, each of which locally stores and maintains data chunks of data files to guarantee the data availability of user data files even when node churn occurs. Further, high-capacity nodes in clusters are selected as backup nodes to achieve adequate load balancing. Simulation results demonstrate the superior effectiveness of the proposed scheme over other file replication schemes. %0 Conference Paper %A Nguyen, Hoai Son %A Nguyen, Dinh Nghia %A Tran, Xuan Hoang %B the 2016 International Conference Advanced Technologies for Communications (ATC) %C Hanoi, Vietnam %D 2016 %F SisLab:1892 %T A Cluster-based File Replication Scheme for DHT-based File Backup Systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1892/ %X DHT-based file backup systems can provide efficient mechanisms for resource placement and lookup. However, these systems must deal with irregular and frequent join/leave of nodes and the problem of load unbalancing between nodes. This paper presents an efficient file replication scheme based on DHT identifier space clustering in order to guarantee data availability and support load balancing for DHT-based file backup systems. The main idea of our method is to divide a DHT network into a number of clusters, each of which locally stores and maintains data chunks of data files to guarantee the data availability for backup data files even in the case of node churn occurred in the network. Further, backup nodes are selected among high-capacity nodes in clusters to achieve a good degree of load balancing. Simulation results demonstrate the effectiveness of our proposed scheme in comparison with other file replication schemes. %0 Conference Paper %A Nguyen, Hoai Son %A Nguyen, Viet Bac %B Hội nghị Quốc gia về Điện tử, Truyền thông và Công nghệ Thông tin %C Ho Chi Minh city, Vietnam %D 2017 %F SisLab:2679 %T Xây Dựng IoT Platform Cho Nhà Thông Minh Tương Thích Chuẩn Echonet Lite %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2679/ %X Để xây dựng các ứng dụng Internet of Things trong nhà thông minh thì các thiết bị cần được kết nối và truy cập một cách dễ dàng dựa trên chuẩn giao thức chung. Trong bài báo này, chúng tôi đề xuất một IoT platform giúp cho các thiết bị trong nhà thông minh có thể giao tiếp với nhau thông qua giao thức Echonet Lite, một giao thức được hỗ trợ bới nhiều công ty sản xuất thiết bị điện dân dụng, đồng thời cho phép các thiết bị trong nhà giao tiếp với các máy tính trên mạng Internet thông qua một Home gateway sử dụng giao thức MQTT (Message Queue Telemetry Transport). IoT platform của chúng tôi cung cấp các chức năng giao tiếp với các thiết bị cũng như lưu trữ thông tin trạng thái và cập nhật trạng thái của các thiết bị cho các ứng dụng IoT trong nhà thông minh. Chúng tôi đã cài đặt thành công IoT platform đề xuất và triển khai thử nghiệm một ứng dụng giám sát và điều khiển thiết bị và một ứng dụng về ổ điện thông minh trên IoT platform này. %0 Journal Article %@ 0866-8612 %A Nguyen, Hoai Son %A Yasuo, Tan %D 2018 %F SisLab:3446 %I VNU Journal of Science %J VNU Journal of Computer Science and Communication Engineering %T Simulation-based Short-term Model Predictive Control for HVAC Systems of Residential Houses %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3446/ %X In this paper, we propose a simple model predictive control (MPC) scheme for Heating, ventilation, and air conditioning (HVAC) systems in residential houses. Our control scheme utilizes fitted thermal simulation to achieve precise prediction of room temperature and energy consumption in each prediction period. The set points of HVAC systems are selected to minimize the amount of energy consumption while maintaining room temperature within a desirable range to satisfy user comfort. Our control system is simple enough to implement in residential houses and is more efficient comparing with rule-based control methods. %0 Journal Article %@ 1947-3532 %A Nguyen, Hoai-Son %A Nguyen, Ngoc Anh %A Bui, Thi Huong Lan %D 2013 %F SisLab:776 %I IGI Global %J International Journal of Distributed Systems and Technologies %N 1 %P 43-55 %T BAM-Chord: DHT-based Bandwidth Adaptive Multicast System %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/776/ %V 4 %0 Journal Article %@ 0916-8516 %A Nguyen, Hoai-Son %A Tan, Yasuo %D 2013 %F SisLab:777 %J IEICE Transactions on Communications %T D-AVTree: A DHT-based Search System Supporting Scalable Multi-Attribute Query %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/777/ %0 Conference Paper %A Nguyen, Hoai-Son %A Tan, Yasuo %A Shinoda, Y. %A VNU-UET, %B 2012 IEEE 26th International Conference on Advanced Information Networking and Applications (AINA) %C Fukuoka, Japan %D 2012 %F SisLab:115 %K Buildings;Indexes;Information services;Load management;Routing;Routing protocols;Search problems;Internet;cryptography;file organisation;information resources;query processing;trees (mathematics);AV tree;DHT-based networks;Internet;distributed attribute-value trees;information resources;multiple-attribute searching;query processing;DHT;Load balancing;Multiple Attribute/Value Searching; %P 795 -802 %T Building Distributed Attribute-Value Trees on DHT-based Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/115/ %X Today, when a huge number of information resources appear in the Internet, one problem arising is how to search and exploit these information resources efficiently and flexibly with high scalability. We believe that the realization of multiple-attribute searching in a distributed manner is an important solution for this problem. In this paper, we propose a solution of building a distributed attribute-value tree (AV tree) on DHT-based networks to realize multiple-attribute searching. Each resource is named by an AV tree and DHT keys are created from each AV branch of a resource name. Resource information is distributed to a DHT-based network by the use of these DHT keys. Our solution keeps the locality of resource names in the key space of DHT networks such that AV branches that share a number of attribute/value pairs (AV pairs) will be mapped to a continuous portion of the key space. Therefore, multiple-attribute searching can be achieved efficiently by querying nodes responsible for a continuous portion of the key space. Our system can achieve both efficiency and a good degree of load balancing even when the distribution of AV pairs in resource names is skewed. Our simulation result shows the efficiency of our solution in respects of query success ratio and the degree of load balancing, comparing with conventional approaches. %0 Conference Paper %A Nguyen, Hoang Anh %A Luu, Viet Hung %A Phan, Anh %A Bui, Quang Hung %A Nguyen, Thi Nhat Thanh %B 6th NAFOSTED Conference on Information and Computer Science %C Hanoi, Vietnam %D 2019 %F SisLab:3583 %T Cau Giay: A Dataset for Very Dense Building Extraction from Google Earth Imagery %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3583/ %X One of the major topics in photogrammetry is the automated extraction of building from data acquired by airborne sensors. What makes this task challenging is the very heterogeneous appearance and dense distribution of buildings in urban areas. While many dataset have been established, none of them pay attention to developing cities where buildings are not well planned. To complement the development of building extraction algorithms, a dataset of high resolution satellite image is constructed in this paper covering Cau Giay district, Hanoi, Vietnam. The dataset consists of 2100 images of size 1024 x 1024 pixels extracted from Google Earth. Shape, size, and construction material differ greatly from building to building, thus make it challenging for state-of-the-art algorithm to accurately extract building location. Some baselines are provided using Convolutional Neural Networks (CNNs). Experimental results show that U-Net model trained with Mean Square Error loss is able to achieve comparable results (OA=92.04). %0 Conference Paper %A Nguyen, Hong Quan %A Nguyen, Phuong Thai %A Dang, Thanh Quyen %A Nguyen, Van Hiep %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1532 %T Automatic Detection of Problematic Rules in Vietnamese Treebank %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1532/ %X Viet Treebank is an annotated corpus newly published in 2009. In this paper, we applied automated methods to detect errors in Viet Treebank based on the concept of equivalence classes proposed by Dickinson. On this basis, we propose an improved method of error detection by transforming the syntax tree based on vertical markovization. Our experimental results on Viet Treebank showed that the scope of error detection extended more than 2 times and improved the precision more than 18.07% in comparison with the base line methods. %0 Conference Paper %A Nguyen, Hong Quan %A Nguyen, Phuong Thai %A Dang, Thanh Quyen %A Nguyen, Van Hiep %B RIVF: 2015 IEEE RIVF International Conference on Computing Communication Technologies - Research, Innovation, and Vision for the Future %C Can Tho, Vietnam %D 2015 %F SisLab:1594 %K Markov processes;equivalence classes;natural language processing;trees (mathematics);Vietnamese Treebank;automatic detection;equivalence classes;problematic rules;syntax trees;vertical Markovization;Abstracts;Accuracy;Buildings;Educational institutions;Natural language processing;Reliability;Syntactics;VTB;ad hoc;auto;detect;equipvalence;error;markovization;treebank %P 13-18 %T Automatic detection of problematic rules in Vietnamese Treebank %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1594/ %X Vietnamese Treebank is a syntactically annotated corpus newly published in 2009. In this paper, we applied automated methods to detect errors in Vietnammese Treebank based on the concept of equivalence classes proposed by Dickinson. On this basis, we propose an improved method of error detection by transforming syntax trees based on vertical markovization. Our experimental results on Vietnamese Treebank showed that the scope of error detection was extended more than 2 times and the precision was improved more than 18.07% in comparison with the base line methods. %0 Conference Paper %A Nguyen, Hong Thinh %A Le, Vu Ha %A Cecile, Barat %A Christophe, Ducottet %B International Joint Conference on Convergence %C Vietnam %D 2016 %F SisLab:2340 %T String Distance For Automatic Image Classification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2340/ %0 Conference Paper %A Nguyen, Hong Thinh %A Nguyen, Viet Dung %A Le, Vu Ha %B International Conference on Advanced Technologies for Communications %C HoChiMinh city, Vietnam %D 2010 %F SisLab:1849 %P 152-157 %T Super-resolution image construction from high-speed camera sequences %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1849/ %0 Conference Paper %A Nguyen, Hong-Quan %A Nguyen, Phuong-Thai %A Dang, Thanh-Quyen %A Nguyen, Van Hiep %B 2015: The 11th IEEE-RIVF International Conference on Computing and Communication Technologies (RIVF 2015) %C Can Tho, Vietnam %D 2015 %F SisLab:1185 %T Automatic Detection of Problematic Rules in Vietnamese Treebank %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1185/ %0 Book %A Nguyen, Hung Son %A Ha, Quang Thuy %A Li, Tianrui %A Małgorzata, Przybyła-Kasperek %B Lecture Notes in Computer Science %D 2018 %E Nguyen, Hung Son %E Ha, Quang Thuy %E Li, Tianrui %E Małgorzata, Przybyła-Kasperek %F SisLab:3157 %I Springer %P 50 %T International Joint Conference on Rough Sets %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3157/ %V 11103 %0 Report %9 Technical Report %@ TR2018-FIT-05 %A Nguyen, Huu Nhat Minh %A Phan, Xuan Hieu %A Tran, Quoc Long %D 2018 %F SisLab:3197 %I VNU University of Engineering and Technology %T Deep learning based detection of vehicles, lane and street sign for behavior cloning in autonomous car %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3197/ %X With the growth of Artificial intelligence and Machine learning, the amount of research for autonomous vehicle is also growing nonstop. But a self-driving system is far too complex with the hardware integration, LiDAR and RADAR involvement, how exactly are these machine learning algorithms being applied, where can a machine learning researcher start to research self-driving car. This report will cover how traditional machine learning methods and state of the are deep learning (semantic segmentation, convolutional neural network - CNN) are being applied to build autonomous cars. The algorithm uses image processing techniques to label lane pixels, classical machine learning and Histogram of Oriented Gradient (HOG) to label vehicle and street sign pixels. Then the data is used to train a semantic segmentation network to extract features for a final CNN to combine with the original image and predict driving command. %0 Conference Paper %A Nguyen, Huu Thuc %A Nguyen, Canh Thuong %A Hoang, Van Xiem %A Byeungwoo, Jeon %B IEEE International Symposium on Broadband Multimedia Systems and Broadcasting 2019 %C Jeju Island, Korea %D 2019 %F SisLab:3450 %P 1-4 %T A Frame Loss Concealment Solution for Spatial Scalable HEVC using Base Layer Motion %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3450/ %X Scalable High Efficiency Video Coding (SHVC), is the most recent video coding solution mainly designed for network adaptive or device adaptive applications. SHVC follows a layered coding structure with one base layer (BL) and one or several enhancement layers (ELs) which can be unequally protected. However, SHVC is often sensitive to the packet loss in unreliable networks, especially in case of ELs. In this paper, we propose a novel error concealment method for the SHVC EL with an assumption that the BL is well protected. First, we recover the partitioning and resample the motion data from collocated BL frame. Following, we remove outliers of motion field by a motion vector refinement algorithm. Lastly, we conceal loss frame by using motion compensation and deblocking filter. Experiments conducted with a rich set of test sequences and for the spatial-scalable SHVC standard have shown that our proposed method significantly outperforms the relevant error concealment methods, e.g., BL Reconstruction Up-sampling (RU) and BL-SKIP in both subjective and objective quality assessments. %0 Conference Paper %A Nguyen, Huu Tung %A Le, Van Giap %A Nguyen, Ngoc Hoa %B SSC: the 2016 Student Scientific Contest %C Hanoi %D 2016 %F SisLab:1502 %T A solution for automatically scanning web vulnerability %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1502/ %0 Conference Paper %A Nguyen, Khoa Hai %A Dang, Duc Hanh %B 2014: International Conference on Advanced Computing and Applications (ACOMP) %C Ho Chi Minh City, Vietnam %D 2014 %F SisLab:1295 %T RTL: A Tool for Precise Model Transformations %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1295/ %0 Journal Article %@ 1859-3585 %A Nguyen, Khoa Hai %A Dang, Duc Hanh %D 2014 %F SisLab:1972 %I Hanoi University of Industry %J Journal of Science and Technology %N 4A %P 230-239 %T RTL: A Tool for Precise Model Transformations %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1972/ %V 52 %0 Conference Paper %A Nguyen, Kim Sao %A Le, Quang Hoa %A Pham, Van At %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1533 %T Một phương pháp thủy vân thuận nghịch mới dựa trên dịch chuyển histogram %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1533/ %X Bài báo đề xuất một phương pháp thủy vân thuận nghịch mới dựa trên dịch chuyển histogram (DCH). Trong khi hầu hết các phương pháp thủy vân dựa trên DCH không có khả năng đóng gói mọi thông tin cần thiết về ảnh gốc vào ảnh thủy vân, nên ở giai đoạn khôi phục chúng phải sử dụng một số thông tin phụ về ảnh gốc, thì phương pháp đề xuất đã khắc phục được nhược điểm trên và không cần sử dụng bất kỳ thông tin phụ nào về ảnh gốc để khôi phục dấu thủy vân cũng như ảnh gốc. Các phương pháp thủy vân như vậy gọi là có tính đóng gói hoặc độc lập thông tin phụ. Giải pháp đóng gói được sử dụng ở đây là kết hợp kỹ thuật chèn bít thấp và dịch chuyển histogram. Ảnh gốc được chia thành hai miền: miền đầu gồm một số ít điểm ảnh được sử dụng để nhúng thông tin phụ bằng kỹ thuật chèn bít thấp, miền thứ hai gồm phần còn lại của ảnh, dùng để nhúng dấu thủy vân theo phương pháp DCH. So sánh với các phương pháp độc lập thông tin phụ khác, phương pháp đề xuất có khả năng nhúng cao hơn và độ phức tạp tính toán thấp hơn. Ưu điểm của giải pháp đóng gói đề xuất là đơn giản, hiệu quả và có thể dễ dàng áp dụng đối với hầu hết các phương pháp thủy vân DCH khác. %0 Journal Article %@ 0218-2130 %A Nguyen, Le Minh %A Phan, Xuan Hieu %A Horiguchi, Susumu %A Shimazu, Akira %D 2007 %F SisLab:2302 %J International Journal on Artificial Intelligence Tools %N 01 %P 129-137 %T A new sentence reduction technique based on a decision tree model %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2302/ %V 16 %0 Conference Paper %A Nguyen, Le Minh %A Shimazu, Akira %A Nguyen, Phuong Thai %A Phan, Xuan Hieu %B Proceedings of the 2007 Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language Learning (EMNLP-CoNLL) %C Prague, Czech Republic %D 2007 %F SisLab:2304 %P 1149-1155 %T A Multilingual Dependency Analysis System Using Online Passive-Aggressive Learning %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2304/ %0 Conference Paper %A Nguyen, Le Minh %A Shimazu, Akira %A Phan, Xuan Hieu %A Nguyen, Phuong Thai %B The 20th IEEE International Conference on Tools with Artificial Intelligence (IEEE ICTAI) %C Dayton, Ohio, USA %D 2008 %F SisLab:2299 %P 135-142 %T Online Structured Learning for Semantic Parsing with Synchronous and lambda-Synchronous Context Free Grammars %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2299/ %0 Journal Article %@ 0165-0114 %A Nguyen, Linh Anh %A Ha, Quang Thuy %A Nguyen, Ngoc Thanh %A Nguyen, Thi Hong Khanh %A Tran, Thanh Luong %A Institute of Informatics, University of Warsaw, Banacha 2, 02-097 Warsaw, Poland, %A aculty of Information Technology, VNU University of Engineering and Technology, 144 Xuan Thuy, Hanoi, Vietnam, %A Department of Information Systems, Faculty of Computer Science and Management, Wroclaw University of Science and Technology, Poland, %A Faculty of Information Technology, Nguyen Tat Thanh University, Ho Chi Minh City, Vietnam, %A Faculty of Information Technology, Electricity Power University, 235 Hoang Quoc Viet, Hanoi, Vietnam, %A Department of Information Technology, University of Sciences, Hue University, 77 Nguyen Hue, Hue city, Vietnam, %D 2019 %F SisLab:3551 %I Elsevier %J Fuzzy Sets and Systems %T Bisimulation and bisimilarity for fuzzy description logics under the Gödel semantics %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3551/ %X Description logics (DLs) are a suitable formalism for representing knowledge about domains in which objects are described not only by attributes but also by binary relations between objects. Fuzzy extensions of DLs can be used for such domains when data and knowledge about them are vague and imprecise. One of the possible ways to specify classes of objects in such domains is to use concepts in fuzzy DLs. As DLs are variants of modal logics, indiscernibility in DLs is characterized by bisimilarity. The bisimilarity relation of an interpretation is the largest auto-bisimulation of that interpretation. In DLs and their fuzzy extensions, such equivalence relations can be used for concept learning. In this paper, we define and study fuzzy bisimulation and bisimilarity for fuzzy DLs under the Gödel semantics, as well as crisp bisimulation and strong bisimilarity for such logics extended with involutive negation. The considered logics are fuzzy extensions of the DL (a variant of PDL) with additional features among inverse roles, nominals, (qualified or unqualified) number restrictions, the universal role, local reflexivity of a role and involutive negation. We formulate and prove results on invariance of concepts under fuzzy (resp. crisp) bisimulation, conditional invariance of fuzzy TBoxes/ABoxes under bisimilarity (resp. strong bisimilarity), and the Hennessy-Milner property of fuzzy (resp. crisp) bisimulation for fuzzy DLs without (resp. with) involutive negation under the Gödel semantics. Apart from these fundamental results, we also provide results on using fuzzy bisimulation to separate the expressive powers of fuzzy DLs, as well as results on using strong bisimilarity to minimize fuzzy interpretations. %0 Journal Article %A Nguyen, Linh Anh %A Ha, Quang Thuy %A Nguyen, Ngoc Thanh %A Nguyen, Thi Hong Khanh %A Tran, Thanh Luong %A Trường Đại học Công nghệ, ĐHQGHN, %D 2020 %F SisLab:4068 %J arXiv.org %T Bisimulation and bisimilarity for fuzzy description logics under the Gödel semantics %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4068/ %X Description logics (DLs) are a suitable formalism for representing knowledge about domains in which objects are described not only by attributes but also by binary relations between objects. Fuzzy extensions of DLs can be used for such domains when data and knowledge about them are vague and imprecise. One of the possible ways to specify classes of objects in such domains is to use concepts in fuzzy DLs. As DLs are variants of modal logics, indiscernibility in DLs is characterized by bisimilarity. The bisimilarity relation of an interpretation is the largest auto-bisimulation of that interpretation. In DLs and their fuzzy extensions, such equivalence relations can be used for concept learning. In this paper, we define and study fuzzy bisimulation and bisimilarity for fuzzy DLs under the Gödel semantics, as well as crisp bisimulation and strong bisimilarity for such logics extended with involutive negation. The considered logics are fuzzy extensions of the DL ALCreg (a variant of PDL) with additional features among inverse roles, nominals, (qualified or unqualified) number restrictions, the universal role, local reflexivity of a role and involutive negation. We formulate and prove results on invariance of concepts under fuzzy (resp. crisp) bisimulation, conditional invariance of fuzzy TBoxex/ABoxes under bisimilarity (resp. strong bisimilarity), and the Hennessy-Milner property of fuzzy (resp. crisp) bisimulation for fuzzy DLs without (resp. with) involutive negation under the Gödel semantics. Apart from these fundamental results, we also provide results on using fuzzy bisimulation to separate the expressive powers of fuzzy DLs, as well as results on using strong bisimilarity to minimize fuzzy interpretations. %0 Conference Paper %A Nguyen, Linh Anh %A Nguyen, Thi Hong Khanh %A Nguyen, Ngoc Thanh %A Ha, Quang Thuy %B 2016 IEEE International Conference on Systems, Man, and Cybernetics (SMC) %D 2016 %F SisLab:2537 %P 004694-004699 %T Bisimilarity for paraconsistent description logics %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2537/ %X We introduce comparisons with respect to information between interpretations in paraconsistent description logics and use them to define bisimilarity for such logics. As bisimilarity is a natural notion for characterizing indiscernibility in modal and description logics, it is useful for concept learning in description logics also when inconsistencies occur. We give preservation results and the Hennessy-Milner property for comparisons with respect to information in paraconsistent description logics. As consequences, we also obtain invariance results and the Hennessy-Milner property for bisimilarity in paraconsistent description logics. %0 Journal Article %@ 1064-1246 %A Nguyen, Linh Anh %A Nguyen, Thi Hong Khanh %A Nguyen, Ngoc Thanh %A Ha, Quang Thuy %A Nguyen, Ngoc Thanh %A Núñez, Manuel %A Trawiński, Bogdan %A Division of Knowledge and System Engineering for ICT, Faculty of Information Technology, Ton Duc Thang University, Ho Chi Minh City, Vietnam E-mail: nguyenanhlinh@tdt.edu.vn, %A Institute of Informatics, University of Warsaw, Warsaw, Poland, %A Faculty of Information Technology, VNU University of Engineering and Technology, Xuan Thuy, Hanoi, Vietnam, %A Electric Power University, Hoang Quoc Viet, Tu Liem, Hanoi, Vietnam, %A Department of Information Systems, Faculty of Computer Science and Management, Wroclaw University of Science and Technology, Wroclaw, Poland, %D 2017 %F SisLab:2473 %J Journal of Intelligent & Fuzzy Systems %N 2 %P 1203-1215 %T Bisimilarity for paraconsistent description logics %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2473/ %V 32 %X We introduce comparisons w.r.t. information between interpretations in paraconsistent description logics and use them to define bisimilarity for such logics. This notion is useful for concept learning in description logics when inconsistencies occur. We give preservation results and the Hennessy-Milner property for comparisons w.r.t. information in paraconsistent description logics. As consequences, we obtain also invariance results and the Hennessy-Milner property for bisimilarity in paraconsistent description logics. %0 Journal Article %@ 1932-4553 %A Nguyen, Linh Trung %A Nguyen, Viet Dung %A Thameri, Messaoud %A Truong, Minh Chinh %A Abed-Meraim, Karim %D 2018 %F SisLab:3317 %I IEEE %J IEEE Journal of Selected Topics in Signal Processing %T Low-complexity adaptive algorithms for robust subspace tracking %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3317/ %X This paper introduces new, low-complexity, adaptive algorithms for robust subspace tracking in certain adverse scenarios of noisy data. First, an adequate weighted least-squares criterion is considered for the design of a robust subspace tracker that is most efficient in the burst noise case. Second, by using data pre-processing and robust statistics estimate, we introduce a second method that is shown to be the most efficient for subspace tracking in the case of impulsive noise (e.g. α-stable noise). Finally, a ‘detect-and-skip’ approach is adopted where the corrupted measurements are detected and treated as ‘missing’ data. The resulting algorithm is particularly effective in the case where the data is affected by sparse ‘outliers’. All these approaches were analyzed and their convergence properties were investigated. Moreover, the proposed subspace tracking algorithms were compared by simulated experiments to some state-of-the-art methods, in different noise/outliers contexts. %Z Early access %0 Conference Paper %A Nguyen, Linh Trung %A Truong, Minh Chinh %A Nguyen, Viet Dung %A Abed-Meraim, Karim %B 12th International Symposium on Medical Information and Communication Technology (ISMICT) %C Sydney, Australia %D 2018 %F SisLab:2947 %P 114-119 %T A Non-Linear Tensor Tracking Algorithm for Analysis of Incomplete Multi-Channel EEG Data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2947/ %X Tensor decomposition is a popular tool to analyse and process data which can be represented by a higher-order tensor structure. In this paper, we consider tensor tracking in challenging situations where the observed data are streaming and incomplete. Specifically, we proposed a non-linear formulation of the PETRELS cost function and based on which we proposed NL-PETRELS subspace and tensor tracking algorithms. The non-linear function allows us to improve the convergence rate. We also illustrated the use of our proposed tensor tracking for incomplete multi-channel electroencephalogram (EEG) data in a real-life experiment in which the data can be represented by a third-order tensor. %0 Conference Paper %A Nguyen, Manh Ha %A Nguyen, Dinh Viet %B Hội thảo Quốc gia lần thứ VII, Một số vấn đề chọn lọc của Công nghệ Thông tin và Truyền thông %C Đà Nẵng, Việt Nam %D 2005 %F SisLab:2991 %P 499-508 %T PERFORMANCE EVALUATION OF RECENT DEVELOPMENT IN TRANSPORT PROTOCOL FOR HIGH BANDWIDTH-DELAY NETWORKS %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2991/ %X The development of gigabit and satellite networking is driving TCP congestion control outside of its natural operating regime. The reason is that AIMD algorithm [2] provides poor utilization of the bottleneck link and long RTT flows are often unable to get a fair allocation of bottleneck bandwidth [3] in those kinds of network. Therefore, this topic is immediately becoming one of universal interests in network performance research community. However, it is difficult to estimate the demand and supply of network resources in dynamic network environment. Although many approaches are used, feedback-based scheme [1], [4] is still the most favorite and prospective one. Some implementations of this approach have been carried out but the detail performance comparison has not been known. This report discusses some advanced approaches and figures out the both advantages and disadvantages of each one; especially BIC-TCP, a new approach based on binary increase congestion control mechanism aimed to solve the TCP friendliness and bandwidth scalability problems. Finally, we inform a number of performance evaluations and parameters which will, in turn, be analyzed after in order to get a better result in various network environments based on ns-2 network simulator. %0 Conference Paper %A Nguyen, Manh Ha %A Nguyen, Dinh Viet %B Hội thảo khoa học Quốc gia lần thứ II: Nghiên cứu cơ bản và Ứng dụng Công nghệ Thông tin" (FAIR2005) %C Thành phố Hồ Chí Minh %D 2006 %F SisLab:2988 %P 239-250 %T PERFORMANCE EVALUATION OF ROUTING PROTOCOLS AND INTERNET ACCESS IN MULTI-HOP MOBILE AD HOC NETWORKS %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2988/ %X A multi-hop mobile ad hoc network is a collection of wireless mobile nodes dynamically forming a temporary network without the usage of any existing network infrastructure or centralized administration. They are self-organized and used in disaster relief, conference, and battlefield environments where infrastructure had been destroyed or does not exist. Because of the unpredictable dynamically moving nature of mobile nodes, the overhead of routing traffic in an ad hoc network is likely to be higher than that in a wire-line network. Consequently,routing protocol have been playing crucial role in the efficiency of overall network operation. There are numerous proposals have been submitted, however many aspects of them needed more investigation and evaluation especially performance and QoS. In this paper, we will evaluate recent development in ad hoc network routing protocols and gateway discovery for Internet access within data intensive environment as well as present quantitative analysis base on simulation by ns-2 network simulator. %0 Journal Article %@ 03772217 %A Nguyen, Minh Anh %A Dang, Giang Thi-Huong %A Hà, Minh Hoàng %A Pham, Minh Trien %D 2021 %F SisLab:4567 %J European Journal of Operational Research %T The min-cost parallel drone scheduling vehicle routing problem %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4567/ %X Adopting unmanned aerial vehicles (UAV), also known as drones, into the last-mile-delivery sector and having them work alongside trucks with the aim of improving service quality and reducing the transportation cost gives rise to a new class of Vehicle Routing Problems (VRPs). In this paper, we introduce a new optimization problem called the min-cost Parallel Drone Scheduling Vehicle Routing Problem (PDSVRP). This problem is a variant of the well-known Parallel Drone Scheduling Traveling Salesman Problem (PDSTSP) recently introduced in the literature in which we allow multiple trucks and consider the objective of minimizing the total transportation costs. We formulate the problem as a Mixed Integer Linear Program and then develop a Ruin and Recreate (R&R) algorithm. Exploiting PDSVRP solution characteristics in an effective manner, our heuristic manages to introduce \sufficient" rooms to a solution via new removal operators during the ruin phase. It is expected to enhance the possibilities for improving solutions later in the recreate phase. Multiple experiments on a new set of randomly generated instances confirm the performance of our approach. To explore the benefits of drone delivery as well as the insight into the impact of related factors on the contribution of drones’ use to operational cost, a sensitivity analysis is conducted. We also adapt the proposed algorithm to solve the PDSTSP and validate it via benchmarks available in the literature. It is shown that our algorithm outperforms state-of-the-art algorithms in terms of solution quality. Out of 90 considered instances, it finds 26 new best known solutions. %0 Conference Paper %A Nguyen, Minh Hai %A Ha, Minh Hoang %A Dinh, Thai Hoang %A Diep, Nguyen %A Dutkiewicz, Eryk %A Tran, The Trung %B CSoNet 2019 %C Ho Chi Minh %D 2019 %F SisLab:3673 %T An Efficient Algorithm for the k-Dominating Set Problem on Very Large-Scale Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3673/ %X The minimum dominating set problem (MDSP) aims to construct the minimum-size subset D⊂V of a graph G=(V,E) such that every vertex has at least one neighbor in D. The problem is proved to be NP-hard [5]. In a recent industrial application, we encountered a more general variant of MDSP that extends the neighborhood relationship as follows: a vertex is a k-neighbor of another if there exists a linking path through no more than k edges between them. This problem is called the minimum k-dominating set problem (MkDSP) and the dominating set is denoted as D_k. The MkDSP can be used to model applications in social networks [2] and design of wireless sensor networks [3]. In our case, a telecommunication company uses the problem model to supervise a large social network up to 17 millions nodes via a dominating subset in which k is set to 3. %0 Journal Article %A Nguyen, Minh Hai %A Ha, Minh Hoang %A Nguyen N, Diep %A Tran, The Trung %D 2020 %F SisLab:4060 %I Springer %J Computational Social Networks %N 7 %T Solving the k-dominating set problem on very large-scale networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4060/ %X The well-known minimum dominating set problem (MDSP) aims to construct the minimum-size subset of vertices in a graph such that every other vertex has at least one neighbor in the subset. In this article, we study a general version of the problem that extends the neighborhood relationship: two vertices are called neighbors of each other if there exists a path through no more than k edges between them. The problem called “minimum k-dominating set problem” (MkDSP) becomes the classical dominating set problem if k is 1 and has important applications in monitoring large-scale social networks. We propose an efficient heuristic algorithm that can handle real-world instances with up to 17 million vertices and 33 million edges. This is the first time such large graphs are solved for the minimum k-dominating set problem %0 Conference Paper %A Nguyen, Minh Hoa %A Vuong, Tung Long %A Le, Thanh Ha %A Nguyen, Thi Thuy %A Nguyen, Do Van %A Nguyen, Dinh Nam %B The 8th International Symposium on Information and Communication Technology (SoICT) %C Nha Trang %D 2017 %F SisLab:2753 %T Moving object detection in compressed domain for high resolution videos %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2753/ %0 Conference Paper %A Nguyen, Minh Nguyet %A Nguyen, Dinh Viet %B Hội thảo Quốc gia lần thứ VII: Một số vấn đề chọn lọc của CNTT và Truyền thông %C Đà Nẵng, Việt Nam %D 2005 %F SisLab:2990 %P 295-305 %T ĐÁNH GIÁ HIỆU SUẤT CÁC GIAO THỨC ĐỊNH TUYẾN TRONG MẠNG KHÔNG DÂY AD HOC %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2990/ %X A mobile ad hoc network (MANET) is a group of wireless mobile nodes dynamically forming a temporary network without the use of any existing network infrastructure or centralized administration. A mobile node acts both as a host and a router forwarding packets to other nodes. Because of the nature of ad hoc networks, there are special demands for ad hoc routing protocols. Certainly, the performance is the most desired issue. This paper presents our efforts in measuring and comparing the performance of ad hoc routing protocols by simulating with the use of NS2 (Network Simulator version 2) package. Three major protocols are considered (DSDV, DSR and AODV). The simulation results show the performance of the protocols and allow us to make conclusions of each one and to compare one with the others. %0 Conference Paper %A Nguyen, Minh Thuan %A Bui, Van Tan %A Vu, Huy Hien %A Nguyen, Phuong Thai %A Luong, Chi Mai %A Department of Computer Science, University of Engineering and Technology, VNU Hanoi, %A Department of Information Technology, University of Economic and Technical Industries, %A Department of Computer Science, University of Engineering and Technology, VNU Hanoi, %A Department of Computer Science, University of Engineering and Technology, VNU Hanoi, %A Department of Language and Speech Processing, Institute of Information Technology, %B International Association of Logopedics and Phoniatrics %D 2018 %F SisLab:3413 %T Enhancing the quality of Phrase-table in Statistical Machine Translation for Less-Common and Low-Resource Languages %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3413/ %X The phrase-table plays an important role in traditional phrase-based statistical machine translation (SMT) system. During translation, a phrase-based SMT system relies heavily on phrase-table to generate outputs. In this paper, we propose two methods for enhancing the quality of phrase-table. The first method is to recompute phrasetable weights by using vector representations similarity. The remaining method is to enrich the phrase-table by integrating new phrase-pairs from an extended dictionary and projections of word vector presentations on the target language space. Our methods produce an attainment of up to 0.21 and 0.44 BLEU scores on in-domain and cross-domain (Asian Language Treebank - ALT) English - Vietnamese datasets respectively. %0 Journal Article %A Nguyen, Minh Thuan %A Nguyen, Phuong Thai %A Nguyen, Van Vinh %A Nguyen Hoang, Minh Cong %D 2020 %F SisLab:4604 %I Association for Computational Linguistics %J Proceedings of the 34th Pacific Asia Conference on Language, Information and Computation %P 207-215 %T Iterative Multilingual Neural Machine Translation for Less-Common and Zero-Resource Language Pairs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4604/ %X Research on providing machine translation systems for unseen language pairs is gaining increasing attention in recent years. However, the quality of their systems is poor for most language pairs, especially for less-common pairs such as Khmer-Vietnamese. In this paper, we show a simple iterative traininggenerating-filtering-training process that utilizes all available pivot parallel data to generate synthetic data for unseen directions. In addition, we propose a filtering method based on word alignments and the longest parallel phrase to filter out noise sentence pairs in the synthetic data. Experiment results on zero-shot Khmer→Vietnamese and Indonesian→Vietnamese directions show that our proposed model outperforms some strong baselines and achieves a promising result under the zero-resource condition on ALT benchmarks. Besides, the results also indicate that our model can easily improve their quality with a small amount of real parallel data. %0 Conference Paper %A Nguyen, Minh Tien %A Ha, Quang Thuy %A Nguyen, Thi Dung %A Nguyen, Tri Thanh %A Nguyen, Le Minh %B KSE: the 2015 International Conference on Knowledge and Systems Engineering %C Ho Chi Minh city, Vietnam %D 2015 %F SisLab:1643 %P 108-113 %T Recognizing Textual Entailment in Vietnamese Text: An Experimental Study %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1643/ %0 Conference Paper %A Nguyen, Minh Tien %A Kitamoto, Asanobu %A Nguyen, Tri Thanh %B 19th Pacific-Asia Conference, PAKDD %C Ho Chi Minh City, Vietnam %D 2015 %F SisLab:1644 %P 64-75 %T TSum4act: A Framework for Retrieving and Summarizing Actionable Tweets During a Disaster for Reaction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1644/ %V 9078 %0 Journal Article %A Nguyen, Minh Tien %A Nguyen, Tri Thanh %A Asanobu, Kitamoto %A Nguyen, Van-Hau %F SisLab:4550 %J International Journal on Artificial Intelligence Tools @ World Scientific Publishing Company %T Exploiting Social Networks as a Live Mass Media Channel During Disasters for Reactions %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4550/ %0 Journal Article %@ 1556-4681 %A Nguyen, Minh Tien %A Tran, Duc Vu %A Nguyen, Le Minh %A Phan, Xuan Hieu %D 2018 %F SisLab:2979 %I ACM %J ACM Transactions on Knowledge Discovery from Data %T Exploiting User Posts for Web Document Summarization %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2979/ %X Relevant user posts such as comments or tweets of a Web document provide additional valuable information to enrich the content of this document. When creating user posts, readers tend to borrow salient words or phrases in sentences. This can be considered as word variation. This paper proposes a framework which models the word variation aspect to enhance the quality of Web document summarization. Technically, the framework consists of two steps: scoring and selection. In the first step, the social information of a Web document such as user posts is exploited to model intra-relations and inter-relations in lexical and semantic levels. These relations are denoted by a mutual reinforcement similarity graph used to score each sentence and user post. After scoring, summaries are extracted by using a ranking approach or concept-based method formulated in the form of Integer Linear Programming. To confirm the efficiency of our framework, sentence and story highlight extraction tasks were taken as a case study on three datasets in two languages, English and Vietnamese. Experimental results show that: (i) the framework can improve ROUGE-scores compared to state-of-the-art baselines of social context summarization and (ii) the combination of the two relations benefits the sentence extraction of single Web documents. %0 Conference Paper %A Nguyen, Minh Trang %A Pham, Van Long %A Nguyen, Chu Chien %A Nguyen, Van Vinh %B ICPRAM 2021 %D 2021 %F SisLab:4603 %P 612-619 %T Object detection and text recognition in large-scale technical drawings %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4603/ %X In this digital transformation era, the demand for automatic pattern extraction from printed materials has never been higher, making it one of the most eminent problems nowadays. In this paper, we propose a new method for pattern recognition in highly complex technical drawings. Our method is a pipeline system that includes two phases: (1) detecting the objects that contain the patterns of interest with improvements of processing large-scale image, and (2) performing character recognition on the objects if they are text patterns with improvements of post-processing task. Our experiments on nearly five thousand real technical drawings show promising results and the capability to reduce manual labeling effort to a great extent. %0 Conference Paper %A Nguyen, Minh Trang %A Tran, Van Lien %A Can, Duy Cat %A Ha, Quang Thuy %A Vu, Thi Ly %A Chng, Eng-Siong %B The 3rd International Conference on Machine Learning and Soft Computing %C Da Lat, Vietnam %D 2019 %F SisLab:3511 %I ACM New York, NY, USA ©2019 %T QASA: Advanced Document Retriever for Open-Domain Question Answering by Learning to Rank Question-Aware Self-Attentive Document Representations %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3511/ %X For information consumers, being able to obtain a short and accurate answer for a query is one of the most desirable features. This motivation, along with the rise of deep learning, has led to a boom in open-domain Question Answering (QA) research. While the problem of machine comprehension has received multiple success with the help of large training corpora and the emergence of attention mechanism, the development of document retrieval in open-domain QA is lagged behind. In this work, we propose a novel encoding method for learning question-aware self-attentive document representations. By applying pair-wise ranking approach to these encodings, we build a Document Retriever, called QASA, which is then integrated with a machine reader to form a complete open-domain QA system. Our system is thoroughly evaluated using QUASAR-T dataset and shows surpassing results compared to other state-of-the-art methods. %0 Journal Article %@ 1752-914X %A Nguyen, Minh-Tien %A Nguyen, Tri Thanh %D 2015 %F SisLab:1470 %I InderScience %J International Journal of Computational Vision and Robotics (IJCVR) %N 3 %P 282-301 %T DESRM: a Disease Extraction System for Real-time Monitoring %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1470/ %V 5 %0 Conference Paper %A Nguyen, Ngoc Bao %A Nguyen, Viet Ha %B ICTACS 2006 %D 2006 %F SisLab:1773 %P 203-214 %T An Approach to Estimate Software Cost within a Narrow Context %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1773/ %0 Journal Article %@ 0916-8532 %A Nguyen, Ngoc Binh %A Pham, Van Huong %A Bui, Ngoc Hai %D 2015 %F SisLab:1475 %I IEICE %J IEICE Transactions on Information and Systems %N 6 %P 1166-1175 %T An Approach to Embedded Software Optimization Based on Reverse Engineering %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1475/ %V E98-D %0 Conference Paper %A Nguyen, Ngoc Binh %A Tran, Xuan Tu %A VNU-UET, %B The 2010 IEICE International Conference on Integrated Circuits Design in Vietnam (ICDV) %C Ho Chi Minh city %D 2010 %F SisLab:55 %T Needs and Challenges in Human Resource Development for Electronic Design in Vietnam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/55/ %X The talk addresses the current status of electronic design (ED) activities in Vietnam, including those in higher education and training programs and in the industry. We also give an overview of ED human resource and R&D activities in Vietnam. The talk then analyzes some challenges and issues to make some proposals and solutions to meet the needs and requirements of ICT and Electronic Industries in Vietnam, including the overseas electronic companies and manufactures and their academic-industrial collaboration with Vietnamese Institutions and Partners. %0 Generic %A Nguyen, Ngoc Hoa %D 2019 %F SisLab:3949 %I MOST %T Nghiên cứu, xây dựng hệ thống đánh giá, quản lý rủi ro và hỗ trợ xử lý sự cố an toàn thông tin trong chính phủ điện tử %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3949/ %0 Patent %@ 1-2021-00205 %A Nguyen, Ngoc Hoa %A Lê Việt Hà, %D 2021 %F SisLab:4461 %T Phương pháp phát hiện đoạn mã độc trong mã nguồn ứng dụng Web sử dụng ngôn ngữ ASP %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4461/ %0 Patent %@ 1-2020-06834 %A Nguyen, Ngoc Hoa %A Lê Việt Hà, %A Phạm Hải Đăng, %D 2021 %F SisLab:4464 %T Phương pháp phát hiện đoạn mã độc trong mã nguồn ứng dụng Web sử dụng ngôn ngữ PHP %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4464/ %0 Patent %@ 1-2021-00205 %A Nguyen, Ngoc Hoa %A Du, Phuong Hanh %D 2021 %F SisLab:4668 %T Phương pháp và hệ thống phát hiện và ngăn chặn xâm nhập mạng sử dụng luật và mô hình học sâu %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4668/ %0 Conference Paper %A Nguyen, Ngoc Hoa %A Le, Viet Ha %A Phung, Van On %A Du, Phuong Hanh %B The Tenth International Symposium on Information and Communication Technology (SoICT 2019) %C Vietnam %D 2019 %F SisLab:3666 %T Toward a Deep Learning Approach for Detecting PHP Webshell %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3666/ %0 Conference Paper %A Nguyen, Ngoc Hoa %A Nguyen, Binh Minh %A Dao, Thanh Chung %A Do, Ba Lam %B 2020 14th International Conference on Computing and Communication Technologies (RIVF) %D 2020 %F SisLab:4006 %T Towards Blockchainizing Land Valuation Certificate Management Procedures in Vietnam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4006/ %0 Conference Paper %A Nguyen, Ngoc Khai %A Truong, Anh Hoang %B SoICT 2017: 8th International Symposium on Information and Communications %C Nha Trang %D 2017 %F SisLab:2712 %T A Compositional Type Systems for Finding Log Memory Bounds of Transactional Programs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2712/ %0 Journal Article %@ 1859-0209 %A Nguyen, Ngoc Khai %A Truong, Anh Hoang %D 2018 %F SisLab:2969 %I Học viện Kỹ thuật Quân sự %J Chuyên san Công nghệ thông tin và Truyền thông (JICT) %T HỆ THỐNG KIỂU ĐỂ TÍNH BỘ NHỚ LOG CỦA CHƯƠNG TRÌNH GIAO DỊCH TỪ BIẾN DÙNG CHUNG %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2969/ %V 11 %X Trong nghiên cứu trước đây, chúng tôi đã xây dựng một hệ thống kiểu để tính tài nguyên tối đa cần sử dụng của một chương trình giao dịch đa luồng. Tuy nhiên, tài nguyên này được tính toán dựa trên các tham số là tài nguyên mà mỗi giao dịch cần sử dụng. Những tham số này do người lập trình phải tự tính toán thủ công dựa trên mã nguồn của chương trình. Vì vậy, kết quả đó vẫn mang tính phương pháp và bán tự động, chưa thuận tiện cho người lập trình. Trong nghiên cứu này, chúng tôi cải tiến hệ thống kiểu để tính được tài nguyên tối đa cần sử dụng của chương trình giao dịch đa luồng một cách hoàn toàn tự động. Người lập trình không cần thực hiện bước tính toán thủ công các tham số như trong nghiên cứu trước. Tài nguyên trong nghiên cứu này được cụ thể hóa là bộ nhớ log của các giao dịch. Để thực hiện được công việc này, ngôn ngữ cũng đã được cải tiến, bổ sung và chúng cũng gần với ngôn ngữ thực tế hơn. %0 Journal Article %@ 2588-1086 %A Nguyen, Ngoc Khai %A Truong, Anh Hoang %A Dang, Duc Hanh %D 2021 %F SisLab:4666 %I VNU %J VNU Journal of Science: Computer Science and Communication Engineering %N 2 %P 1-19 %T Estimate the memory bounds required by shared variables in software transactional memory programs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4666/ %V 37 %0 Journal Article %@ 2588-1086 %A Nguyen, Ngoc Khai %A Truong, Anh Hoang %A Dang, Duc Hanh %F SisLab:4608 %I VNU %J VNU Journal of Science: Computer Science and Communication Engineering %T Estimate the memory bounds required by shared variables in software transactional memory programs", ID 280. %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4608/ %0 Conference Paper %A Nguyen, Ngoc Khai %A Truong, Anh Hoang %A Dang, Van Hung %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1530 %T A type system for calculating the maximum log memory used by transactional programs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1530/ %X We developed a type system for estimating an upper bound of memory that multi-threaded and nested transactional programs may require for their transaction logs. In our previous works, we only esti- mated the maximum number of logs that can coexist by static type and effect systems. This work extends our previous language that allows ones to specify also the size of logs in a transaction, and then develop a type system that can infer the memory bound required by the transaction logs. %0 Conference Paper %A Nguyen, Ngoc Khuong %A Le, Anh Cuong %A Nguyen, Viet Ha %B IUKM 2018 %D 2018 %F SisLab:3041 %P 166-178 %T An Attention-Based Long-Short-Term-Memory Model for Paraphrase Generation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3041/ %X Neural network based sequence-to-sequence models have shown to be the effective approach for paraphrase generation. In the problem of paraphrase generation, there are some words which should be ignored in the target text generation. The current models do not pay enough attention to this problem. To overcome this limitation, in this paper we propose a new model which is a penalty coefficient attention-based Residual Long-Short-Term-Memory (PCA-RLSTM) neural network for forming an end-to-end paraphrase generation model. Extensive experiments on the two most popular corpora (PPDB and WikiAnswers) show that our proposed model’s performance is better than the state-of-the-art models for paragraph generation problem. %0 Conference Paper %A Nguyen, Ngoc Khuong %A Le, Anh Cuong %A Nguyen, Viet Ha %B 2018 12th Multi-disciplinary International Conference on Artificial Intelligence (MIWAI) %C Hanoi %D 2018 %F SisLab:3353 %P 161-175 %T A Hierarchical Conditional Attention-Based Neural Networks for Paraphrase Generation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3353/ %0 Conference Paper %A Nguyen, Ngoc Khuong %A Pham, Duc Hong %A Le, Anh Cuong %A Pham, Hong Thai %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1534 %T Improving Maximum Entropy Part-of-Speech Tagger with Transformation Based Learning Model %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1534/ %X As well know, part-of-Speech (POS) tagging is basic and central problem of natural language processing branch. Development of a POS tagger will influence several pipelined modules of natural language understanding system including information extraction and retrieval; machine translation; partial parsing and word sense disambiguation. In recent years, there has been a growing interest in data-driven machine learning disambiguation methods for POS tagging with results be very close state-of-art of this problem. Improvement performance of these methods has been posed challenges for researchers. In this paper, we use one of the best machine learning method (Maximum Entropy Model - MEM) to do automatic annotation of part-of-speech in basedline then improve accuracy by using Transformation Based Learning (TBL). Our approach based on an incremental knowledge acquisition method where rules are stored in an exception structure and new rules are only selected to correct the errors are positive on courpus. Experimental results on English Peen Treebank show that our method greatly improve accuracy than the naive baseline model with state-of-the-art accuracy(97.14%). Special, we also do experiments on Vietnamese Viet TreeBank corpus and experimental results show that the proposed Vietnamese POS tagging system outperforms the other state-of-the-art Vietnamese taggers with 93.50% overall accuracy. %0 Conference Paper %A Nguyen, Ngoc Tuan %A Nguyen, Thi Nhat Thanh %A Le, Thanh Ha %B 2013: International Conference on Electronics, Information and Communication (ICEIC) %D 2013 %F SisLab:787 %P 201-203 %T Text Recognition Method for Trademark Images %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/787/ %0 Conference Paper %A Nguyen, Ngoc Vu %A Bui, Hong Son %A Ha, Quang Thuy %B 5th EAI International Conference, INISCOM 2019 %C Ho Chi Minh City, Vietnam %D 2019 %F SisLab:3575 %T Ontology-Based Semantic Search for National Database of Natural Resources and Environment %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3575/ %X Semantic search helps the user queries to be understandable for electric agents searching. In this way, ontology plays the main role to define the semantic and the relations between user queries. The national database of natural resources and environment is a very large database system. Therefore, building a search engine software for the system with high accuracy and fast speed is very important for sharing information in the field of natural resources and environment. However, the existing search engine software in the national database needs to be improved to better meet user’s needs. We proposed the architecture of ontology-based semantic search for the national database of natural resources and environment. Based on the proposed architecture, we have built semantic search software (VnNRESS) to demonstrate better results than the existing search software (NRESearch). %0 Conference Paper %A Nguyen, Ngoc Vu %A Nguyen, Hai Chau %A Ha, Quang Thuy %B Proceedings of 6th NAFOSTED Con- ference on Information and Computer Science (NICS) %C Hanoi, Vietnam %D 2019 %F SisLab:4235 %P 12-13 %T Developing a Domain Ontology for Natural Resources and Environment %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4235/ %X Semantic search helps the user queries to be understandable for electric agents searching. In this way, ontology plays the main role to define the semantic and the relations between user queries. Ontologies provide features like a common vocabulary, reusability, machine-readable content, and also allows for semantic search, facilitate agent interaction and ordering and structuring of knowledge for the Semantic Web (Web 3.0) application. Ministry of Natural Resources and Environment is currently managing a large amount of data for 9 areas: land, mineral geology, water resources, environment, seas and islands, hydrometeorology, climate change, survey and mapping and remote sensing. Therefore, the need for data management, data integration and semantic search is essential and developing a domain ontology for natural resources and environment helps to solve these problems. In this paper, we propose a methodology for constructing a domain ontology for natural resources and environment based on integrating some existing ontologies (in English) and extending them using reliable resources in Vietnam. The ontology has been built is an important input for further researches on ontology learning, natural language processing, automated question answering and semantic search. %0 Conference Paper %A Nguyen, Ngoc Vu %A Nguyen, Thi Lan %A Nguyen, Thi Cam Van %A Tran, Mai Vu %A Ha, Quang Thuy %B ACIIDS 2019: Intelligent Information and Database Systems %C Yogyakarta, Indonesia %D 2019 %F SisLab:3471 %P 90-102 %T A Character-Level Deep Lifelong Learning Model for Named Entity Recognition in Vietnamese Text %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3471/ %X Lifelong Machine Learning (LML) is a continuous learning process, in which the knowledge learned from previous tasks is accumulated in the knowledge base, then the knowledge will be used to support future learning tasks, for which it may be only a few of samples exists. However, there is a little of studies on LML based on deep neural networks for Named Entity Recognition (NER), especial in Vietnamese. We propose DeepLML-NER model, a lifelong learning model based on using deep learning methods with a CRFs layer, for NER in Vietnamese text. DeepLML-NER includes an algorithm to extract the knowledge of “prefix-features” of Named Entities in previous domains. Then the model uses the knowledge stored in the knowledge base to solve a new NER task. The effect of the model was demonstrated by in-domain and cross-domain experiments, achieving promising results. %0 Journal Article %@ 2196-8888 %A Nguyen, Ngoc Vu %A Nguyen, Thi Lan %A Nguyen, Thi Cam Van %A Tran, Mai Vu %A Nguyen, Tri Thanh %A Ha, Quang Thuy %D 2019 %F SisLab:3590 %I Springer %J Vietnam J. Computer Science %N 4 %P 471-487 %T Improving Named Entity Recognition in Vietnamese Texts by a Character-Level Deep Lifelong Learning Model %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3590/ %V 6 %X Named entity recognition (NER) is a fundamental task which affects the performance of its dependent task, e.g. machine translation. Lifelong machine learning (LML) is a continuous learning process, in which the knowledge base accumulated from previous tasks will be used to improve future learning tasks having few samples. Since there are a few studies on LML based on deep neural networks for NER, especially in Vietnamese, we propose a lifelong learning model based on deep learning with a CRFs layer, named DeepLML–NER, for NER in Vietnamese texts. DeepLML–NER includes an algorithm to extract the knowledge of “prefix-features” of named entities in previous domains. Then the model uses the knowledge in the knowledge base to solve the current NER task. Preprocessing and model parameter tuning are also investigated to improve the performance. The effect of the model was demonstrated by in-domain and cross-domain experiments, achieving promising results. %0 Conference Paper %A Nguyen, Ngoc Vu %A Tran, Mai Vu %A Nguyen, Hai Chau %A Ha, Quang Thuy %A Trường Đại học Công nghệ, ĐHQGHN, %A Cục CNTT, Bộ Tài nguyên và Mội trường, %B 11th International Conference on Information Science and Applications 2020 (ICISA2020) %C Virtual Conference %D 2020 %F SisLab:4175 %T A Word + Character Embedding based Relation Extraction Frame for Domain Ontology of Natural Resources and Environment %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4175/ %X Building domain ontology is a challenging problem, and there are many different approaches for domain ontology construction. However, most of these approaches are still mainly using manual methods [1]. Ontology enrichment is a fairly standard approach in domain ontology construction, in which semi-automated methods and automated methods of ontology learning from a derived ontology. Relation extraction is one of the ways for ontology enrichment. Relation extraction techniques include law-based techniques, machine learning-based techniques with three typical methods: supervised learning, semi-supervised learning, and unsupervised learning. This paper proposes a word + character embedding-based relation extraction frame for the Vietnamese domain ontology of natural resources and environment. The model's effect was demonstrated by experiments in the domain of natural resources and the envi-ronment and achieving promising results. %0 Conference Paper %A Nguyen, Pham Xuan %A Le, Quang Hieu %B 2013: The Fifth International Conference on Knowledge and Systems Engineering (KSE 2013) %C Hanoi, Vietnam %D 2013 %F SisLab:757 %P 261-270 %T A New Improved Term Weighting Scheme for Text Categorization %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/757/ %V 1 %0 Journal Article %@ 1574-0218 %A Nguyen, Phuong Thai %A Le, Anh Cuong %A Ho, Tu Bao %A Nguyen, Van Hiep %A VNU-UET, %A JAIST, %D 2015 %F SisLab:1467 %I Springer %J Language Resources and Evaluation %N 3 %P 487-519 %T Vietnamese Treebank Construction and Entropy-based Error Detection %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1467/ %V 49 %0 Journal Article %@ 1574-0218 %A Nguyen, Phuong Thai %A Le, Anh Cuong %A Ho, Tu Bao %A Nguyen, Van Hiep %D 2015 %F SisLab:1592 %I Springer %J Language Resources and Evaluation %N 3 %P 487-519 %T Vietnamese treebank construction and entropy-based error detection %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1592/ %V 49 %X Treebanks, especially the Penn treebank for natural language processing (NLP) in English, play an essential role in both research into and the application of NLP. However, many languages still lack treebanks and building a treebank can be very complicated and difficult. This work has a twofold objective. Firstly, to share our results in constructing a large Vietnamese treebank (VTB) with three levels of annotation including word segmentation, part-of-speech tagging, and syntactic analysis. Major steps in the treebank construction process are described with particular regard to specific Vietnamese properties such as lack of word delimiter and isolation. Those properties make sentences highly syntactically ambiguous, and therefore it is difficult to ensure a high level of agreement among annotators. Various studies of Vietnamese syntax were employed not only to define annotations but also to systematically deal with ambiguities. Annotators were supported by automatic labelling tools, which are based on statistical machine learning methods, for sentence pre-processing and a tree editor for supporting manual annotation. As a result, an annotation agreement of around 90 % was achieved. Our second objective is to present our method for automatically finding errors and inconsistencies in treebank corpora and its application to the construction of the VTB. This method employs the Shannon entropy measure in a manner that the more reduced entropy the more corrected errors in a treebank. The method ranks error candidates by using a scoring function based on conditional entropy. Our experiments showed that this method detected high-error-density subsets of original error candidate sets, and that the corpus entropy was significantly reduced after error correction. The size of these subsets was only about one third of the whole set, while these subsets contained 80–90 % of the total errors. This method can also be applied to languages similar to Vietnamese. %0 Journal Article %A Nguyen, Phuong Thai %A Pham, Van Lam %A Nguyen, Hoang Trung %A Tran, Ngoc Anh %A Truong, Thi Thu Ha %A Nguyen, Van Hiep %D 2014 %F SisLab:1184 %J Tạp chí Từ điển học và Bách khoa thư %T Tổng quan về xây dựng mạng từ tiếng Việt %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1184/ %0 Conference Paper %A Nguyen, Quang Huy %A Nguyen, Van Vinh %A Nguyen, Phu Binh %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1535 %T Nghiên cứu về gán nhãn vai nghĩa cho tiếng Việt %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1535/ %X Gán nhãn vai nghĩa (Semantic Role Labeling - SRL) là bài toán xác định vai nghĩa cho các thành phần trong câu, làm rõ cấu trúc, mối quan hệ giữa vị từ và các tham đối cấu thành câu. Về bản chất, gán nhãn vai nghĩa chỉ là một bài toán trung chuyển, chứ không thật sự là nền tảng của một ứng dụng xử lý ngôn ngữ tự nhiên cụ thể nào. Tuy nhiên, đây lại là một bước cần thiết không thể bỏ qua trong các ứng dụng liên quan đến ngôn ngữ tự nhiên. Việc xác định đúng vai nghĩa của các thành phần trong câu là một vấn đề trung tâm của mọi hệ xử lý ngôn ngữ tự nhiên. Gán nhãn vai nghĩa thực chất là một dạng phân tích ngữ nghĩa nông, xác định và gán nhãn vai nghĩa cho các thành phần trong câu đối với một vị tố cụ thể, thông thường là động từ. Trong các nghiên cứu của mình, chúng tôi tiến hành định nghĩa một số vai nghĩa thông dụng, xây dựng kho ngữ liệu tiếng Việt có gán nhãn vai nghĩa. Bước đầu áp dụng phương pháp phân tích cú pháp đầy đủ để xây dựng công cụ gán nhãn vai nghĩa tự động cho tiếng Việt. Các thử nghiệm với tiếng Việt cho kết quả tương đối khả quan và đáng tin cậy. %0 Conference Paper %A Nguyen, Quang Trung %A Bui, The Duy %A Ma, Thi Chau %B NICS: 2nd National Foundation for Science and Technology Development Conference on Information and Computer Science %C Ho Chi Minh city, Vietnam %D 2015 %F SisLab:1464 %K Bayes methods;signal classification;speech recognition;transforms;vectors;ISOLET;LNBNN;SIFT features;digits;feature vectors;image based approach;local naive Bayes nearest neighbor;places databases;scale-invariant feature transform;speech perception;speech signal classification;spoken word recognition;training data;training phase;Databases;Feature extraction;Hidden Markov models;Mel frequency cepstral coefficient;Speech;Speech recognition;Training;lnbnn;sift;speech classification;speech perception %P 208-213 %T An Image based Approach for Speech Perception %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1464/ %X Classification of speech signal is one of the most vital problems in speech perception and spoken word recognition. Although, there have been many studies on the classification of speech signals but the results are still limited. In this paper, we propose an image based approach for speech signal classification based on the combination of Local Naïve Bayes Nearest Neighbor (LNBNN) and Scale-invariant Feature Transform (SIFT) features. The proposed approach allows training feature vectors to have different sizes and no re-training is needed for additional training data after training phase. With this approach, achieved classification results are very satisfactory. They are 72.8, 100 and 95.0 on the ISOLET, Digits and Places databases, respectively. %0 Conference Paper %A Nguyen, Quang-Trung %A Pham, Ngoc Hung %B 2013: 2nd International Conference on Context-Aware Systems and Applications (ICCASA 2013) %C Phu Quoc, Vietnam %D 2013 %F SisLab:770 %T An Efficient Method for Automated Control Flow Testing of Programs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/770/ %0 Journal Article %@ 1570-0844 %A Nguyen, Quoc Dat %A Nguyen, Quoc Dai %A Pham, Bao Son %D 2015 %F SisLab:1491 %I IOS Press %J Semantic Web Journal %T Ripple Down Rules for Question Answering %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1491/ %X Recent years have witnessed a new trend on building ontology-based question answering systems, that is to use semantic web information to provide more precise answers to users' queries. However, these systems are mostly designed for English, therefore, we introduce in this paper such a system for Vietnamese, that is, to the best of our knowledge, the first one made for Vietnamese. Different from most of previous works, we propose an approach that systematically builds a knowledge base of grammar rules for processing each input question into an intermediate representation element. Then we take this element with respect to a target ontology by applying concept-matching techniques for returning an answer. Experimental results show that the performance of the system on a wide range of Vietnamese questions is promising with accuracies of 84.1% and 82.4% for analyzing question and retrieving answer, respectively. Furthermore, our approach to the question analysis can easily be applied to new domains and new languages, thus saving time and human effort. %0 Journal Article %A Nguyen, Quoc Dat %A Nguyen, Quoc Dai %A Pham, Duc Dang %A Pham, Bao Son %D 2016 %F SisLab:2254 %J AI Communications %N 3 %P 409-422 %T A robust transformation-based learning approach using ripple down rules for part-of-speech tagging %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2254/ %V 29 %0 Conference Paper %A Nguyen, Quoc Huy %A Pham, Huu Bang %A Nguyen, Thi Nhat Thanh %A Pham, Van Cu %A Bui, Quang Hung %B 2014: Hội thảo ứng dụng GIS toàn quốc %D 2014 %F SisLab:1423 %P 855-863 %T Xây dựng hệ thống WebGIS hỗ trợ quản lý và chia sẻ dữ liệu ảnh viễn thám mã nguồn mở: Áp dụng cho dữ liệu ảnh MODIS %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1423/ %V 2 %0 Conference Paper %A Nguyen, Son %A Vo, Hieu %A Pham, Ngoc Hung %B The 2015 Symposium on Information and Communication Technology %C Hue, Vietnam %D 2015 %F SisLab:1742 %T A Correlation-aware Negotiation Approach for Service Composition %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1742/ %0 Conference Paper %A Nguyen, Son Hoang %A Le, Sy Vinh %A Le, Si Quang %A VNU-UET, %B 2012 Fourth International Conference on Knowledge and Systems Engineering (KSE) %C Danang, Vietnam %D 2012 %F SisLab:105 %K Accuracy;Approximation methods;Entropy;Gaussian distribution;Genomics;Humans;biology computing;cellular biophysics;genomics;lab-on-a-chip;Gencall;GenoSNP;Illumina BeadChips;Jeffreys distance;Kenyan people;bad SNP detection;chromosome;human genomes;microarray chips;probe intensities data;single nucleotide polymorphisms;software Illuminus;unstable SNP detection;SNP genotype;bad SNPs;quality control;relative entropy; %P 18 -25 %T Detecting Bad SNPs from Illumina BeadChips Using Jeffreys Distance %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/105/ %X Current microarray technologies are able to assay thousands of samples over million of SNPs simultaneously. Computational approaches have been developed to analyse a huge amount of data from microarray chips to understand sophisticated human genomes. The data from microarray chips might contain errors due to bad samples or bad SNPs. In this paper, we propose a method to detect bad SNPs from the probe intensities data of Illumina Beadchips. This approach measures the difference among results determined by three software Illuminus, GenoSNP and Gencall to detect the unstable SNPs. Experiment with SNP data in chromosome 20 of Kenyan people demonstrates the usefulness of our method. This approach reduces the number of SNPs that are needed to check manually. Furthermore, it has the ability in detecting bad SNPs that have not been recognized by other criteria. %0 Journal Article %@ 2010-3743 %A Nguyen, Thac Huy %A Do, The Chuan %A Nguyen, Viet Anh %D 2016 %F SisLab:1747 %J International Journal of Computer and Communication Engineering %N 4 %P 276-285 %T Implicit Feedback Mechanism to Manage User Profile Applied in Vietnamese News Recommender System %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1747/ %V 5 %X The appearance of a vast array of online newspapers makes appropriate news recommendation for users become critically important. A number of researches in this area have been conducted in recent years. This paper presents the Vietnamese content-based news recommender system. Based on information about news stories that users have read and given feedback, the system can automatically determine news stories that users may want to read later. The center of the system is a hybrid user profile which can diversely model news reading features of individual users thanks to a combination of short-term and long-term information models along with self-description. The paper also introduces and evaluates a mechanism for implicit feedback. After installing the system and conducting several experiments, as well as collecting feedback from users, it is shown that the system is relatively adaptable to users and can recommend appropriate news with a high degree of accuracy, which significantly increases the effectiveness of gathering information compared to traditional news reading. %0 Report %9 Technical Report %A Nguyen, Thanh Cong %A Nguyen, Dinh Tuan %A Tran, Quoc Long %D 2018 %F SisLab:3281 %T INFORMATION EXTRACTION FROM ID CARD VIA COMPUTER VISION TECHNIQUES %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3281/ %X An identity card is a type of identify documents of each Vietnamese citizen to prove his/her identity. In this modern information technology era, automatically extracting information from identity cards is applied in telecommunications, healthcare, hotels, and more. This report works on the use of computer vision techniques to build a system which extracts information from identity cards. There are three main steps: Identifying identity cards, locating information and extracting information. In particular, identifying indentity cards and locating information are accomplished by basic image processing techniques to increase the speed of the entire system. At the information extraction step, the achievements in text recognition on complex background are used to increase accuracy. After development and experimentation, the system has achieved a relatively high degree of accuracy. The accuracy of the whole system is 95.28%, which makes it possible to be put into practice %0 Journal Article %@ 2169-3536 %A Nguyen, Thanh Cong %A Yuris Mulya, Saputra %A Huynh, Van Nguyen %A Nguyen, Ngoc Tan %A Tran, Viet Khoa %A Bui Minh, Tuan %A Diep, Nguyen %A Dinh, Thai Hoang %A Vu, Xuan Thang %A Eryk, Dutkiewicz %A Symeon, Chatzinotas %A Bjorn, Ottersten %F SisLab:4333 %I IEEE %J IEEE Access %T A Comprehensive Survey of Enabling and Emerging Technologies for Social Distancing——Part II: Emerging Technologies and Open Issues %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4333/ %X This two-part paper aims to provide a comprehensive survey on how emerging technologies, e.g., wireless and networking, artificial intelligence (AI) can enable, encourage, and even enforce social distancing practice. In Part I, an extensive background of social distancing is provided, and enabling wireless technologies are thoroughly surveyed. In this Part II, emerging technologies such as machine learning, computer vision, thermal, ultrasound, etc., are introduced. These technologies open many new solutions and directions to deal with problems in social distancing, e.g., symptom prediction, detection and monitoring quarantined people, and contact tracing. Finally, we discuss open issues and challenges (e.g., privacy-preserving, scheduling, and incentive mechanisms) in implementing social distancing in practice. As an example, instead of reacting with ad-hoc responses to COVID-19-like pandemics in the future, smart infrastructures (e.g., next-generation wireless systems like 6G, smart home/building, smart city, intelligent transportation systems) should incorporate a pandemic mode in their standard architectures/designs. %0 Journal Article %@ 2169-3536 %A Nguyen, Thanh Cong %A Yuris Mulya, Saputra %A Nguyen, Ngoc Tan %A Tran, Viet Khoa %A Bui, Minh Tuan %A Diep, Nguyen %A Dinh, Thai Hoang %A Vu, Xuan Thang %A Eryk, Dutkiewicz %A Symeon, Chatzinotas %A Bjorn, Ottersten %D 2020 %F SisLab:4332 %I IEEE %J IEEE Access %P 153479-153507 %T A Comprehensive Survey of Enabling and Emerging Technologies for Social Distancing—Part I: Fundamentals and Enabling Technologies %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4332/ %V 8 %X Social distancing plays a pivotal role in preventing the spread of viral diseases illnesses such as COVID-19. By minimizing the close physical contact among people, we can reduce the chances of catching the virus and spreading it across the community. This two-part paper aims to provide a comprehensive survey on how emerging technologies, e.g., wireless and networking, artificial intelligence (AI) can enable, encourage, and even enforce social distancing practice. In this Part I, we provide a comprehensive background of social distancing including basic concepts, measurements, models, and propose various practical social distancing scenarios. We then discuss enabling wireless technologies which are especially effectin social distancing, e.g., symptom prediction, detection and monitoring quarantined people, and contact tracing. The companion paper Part II surveys other emerging and related technologies, such as machine learning, computer vision, thermal, ultrasound, etc., and discusses open issues and challenges (e.g., privacypreserving, scheduling, and incentive mechanisms) in implementing social distancing in practice. %0 Report %9 Technical Report %@ UET-FIT-2018xyz %A Nguyen, Thanh Ngan %A Nguyen, Minh Trang %A Dang, Thanh Hai %D 2018 %F SisLab:3200 %T Disease Named Entity Normalization Using Pairwise Learning To Rank and Deep Learning %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3200/ %0 Journal Article %@ 0866-7756 %A Nguyen, Thanh Thuy %A Ha, Quang Thuy %A Phan, Xuan Hieu %A Nguyen, Tri Thanh %D 2018 %F SisLab:3161 %J Tạp chí Công Thương - Các kết quả nghiên cứu khoa học và Ứng dụng công nghệ %T Trí tuệ nhân tạo trong thời đại số: Bối cảnh thế giới và liên hệ với Việt Nam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3161/ %X Trí tuệ nhân tạo hiện đang phát triển với tốc độ “hàm mũ”, có nhiều đóng góp quan trọng vào sản xuất, kinh doanh, dịch vụ và đời sống con người. Tuy nhiên, Trí tuệ nhân tạo là một lĩnh vực rất phức tạp và cũng tạo ra nhiều thách thức rất đáng lo ngại. Hiểu biết đúng về Trí tuệ nhân tạo để nắm bắt đúng và kịp thời các cơ hội và thách thức từ Trí tuệ nhân tạo là rất cần thiết đối với mỗi con người, mỗi tổ chức và mỗi quốc gia. Bài viết này cung cấp một khái quát chung về Trí tuệ nhân tạo, sự phát triển vượt bậc của Trí tuệ nhân tạo trong thời đại số và các thách thức lớn từ Trí tuệ nhân tạo. Bài viết cũng đề cập tới tình hình nghiên cứu, triển khai Trí tuệ nhân tạo tại Việt Nam. Trên cơ sở khảo sát nội dung cơ bản các chiến lược phát triển Trí tuệ nhân tạo quốc gia của một số nước trên thế giới, chúng tôi có một vài trao đổi về chiến lược phát triển Trí tuệ nhân tạo quốc gia của Việt Nam. Xây dựng và thực hiện thành công chiến lược phát triển Trí tuệ nhân tạo quốc gia sẽ góp phần tạo động lực cho sự nghiệp phát triển kinh tế - xã hội của Việt Nam. %0 Journal Article %@ 1859-387X %A Nguyen, Thanh Trung %A Trinh, Dinh Hoan %A Nguyen, Linh Trung %A Marie, Luong %D 2019 %F SisLab:3697 %I REV %J REV Journal on Electronics and Communications %N 3-4 %P 78-88 %T Low-dose CT image denoising using image decomposition and sparse representation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3697/ %V 9 %X X-ray computed tomography (CT) is now a widely used imaging modality for numerous medical purposes. The risk of high X-ray radiation may induce genetic, cancerous and other diseases, demanding the development of new image processing methods that are able to enhance the quality of low-dose CT images. However, lowering the radiation dose increases the noise in acquired images and hence affects important diagnostic information. This paper contributes an efficient denoising method for low-dose CT images. A noisy image is decomposed into three component images of low, medium and high frequency bands; noise is mainly presented in the medium and high component images. Then, by exploiting the fact that a small image patch of the noisy image can be approximated by a linear combination of several elements in a given dictionary of noise-free image patches generated from noise-free images taken at nearly the same position with the noisy image, noise in these medium and high component images are effectively eliminated. Specifically, we give new solutions for image decomposition to easily control the filter parameters, for dictionary construction to improve the effectiveness and reduce the running-time. Instead of using a large dataset of patches, only a structured small part of patches extracted from the raw data is used to form a dictionary, to be used in sparse coding. In addition, we illustrate the effectiveness of the proposed method in preserving image details which are subtle but clinically important. Experimental results conducted on both synthetic and real noise data demonstrate that the proposed method is competitive with the state-of-the-art methods. %0 Book Section %A Nguyen, Thanh Tung %A Nguyen, Van Duc %A Nguyen, Hai Thanh %A Phan, Cong Vinh %A Nguyen, Dai Tho %A International School, Vietnam National University, Hanoi, %A Hanoi University of Technology, %A Nguyen Tat Thanh University, %A University of Engineering and Technology, Vietnam National University, Hanoi, %B Proceedings of the International Conference on Context-Aware Systems and Applications (ICCASA 2012), %C Berlin %D 2013 %F SisLab:2374 %I Springer %P 183-191 %S Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering %T Power Save Protocol Using Chain Based Routing %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2374/ %V 109 %X Sensor networks are deployed in numerous military and civil applications, such as remote target detection, weather monitoring, weather forecast, natural resource exploration and disaster management. Despite having many potential applications, wireless sensor networks still face a number of challenges due to their particular characteristics that other wireless networks, like cellular networks or mobile ad hoc networks do not have. The most difficult challenge of the design of wireless sensor networks is the limited energy resource of the battery of the sensors. This limited resource restricts the operational time that wireless sensor networks can function in their applications. Routing protocols play a major part in the energy efficiency of wireless sensor networks because data communication dissipates most of the energy resource of the networks. In many situations, a base station only needs a summary of the gathered information. For example, the base station might only require the maximum temperature of all sub-regions, each covered by a sensor or the average temperature of all sensors in the network. For similar types of application, data aggregation can be applied at all sensor nodes before the data is forwarded to the base station. The above discussions imply a new family of protocols called chain-based protocols. In the protocols, all sensor nodes sense and gather data in an energy efficient manner by cooperating with their closest neighbors. The gathering process can be done until an elected node calculates the final data and sends the data to the base station. %0 Conference Paper %A Nguyen, ThanhDat %A Kifor, Claudiu Vasile %A Nguyen, Hoai Son %B 6th IEEE International Conference on Communications and Electronics (ICCE) %C Ha Long, Vietnam %D 2016 %F SisLab:1977 %P 7-12 %T DSMAV: An improved solution for multi-attribute search based on load capacities %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1977/ %X DHT (Distributed Hash Table) such as CHORD or PARTRY facilitates information searching in scalable systems. Two popular DHT-based approaches for range or multi-attribute search are to rely on attribute-value tree and a combination of attributes and values. However, tradeoff between a load balancing mechanism and query efficiency is a challenging task for such information searching systems. In this paper, we propose improved algorithms for a system called DSMAV in which information resources are distributed fairly among nodes and found based on multi-attribute queries in a small number of hop counts. Our system creates identifiers from resource names, each of which is a combination of attribute-value pairs (AV pairs). In order to overcome problem of load imbalance among nodes because of frequent appearance of AV pairs in many resource names, we control the quantity of information resources distributed to a node based on load capacity of each node. Moreover, periodical updating processes on a node enable our load balancing mechanism to be maintained accurately. As a result, our improvement evaluated by simulated results depicts a good degree of load balance as well as query efficiency. %0 Conference Paper %A Nguyen, Thao %A Le, Sy Vinh %B ICBBB %D 2019 %F SisLab:3526 %P 36-42 %T A Hybrid Approach to Optimize the Number of Recombinations in Ancestral Recombination Graphs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3526/ %X Building ancestral recombination graphs (ARG) with minimum number of recombination events for large datasets is a challenging problem. We have proposed ARG4WG and REARG heuristic algorithm for constructing ARGs with thousands of whole genome sequences. However, these algorithms do not result in ARGs with minimal number of recombination events. In this work, we propose GAMARG algorithm, an improvement of ARG4WG, to optimize the number of recombination events in ARG building process. Experiment with different datasets showed that GAMARG algorithm outperforms other heuristic algorithms in building ARGs for large datasets. It also is much better than other heuristic algorithms and comparable to exhaustive search methods for small datasets. %0 Journal Article %@ 1545-5963 %A Nguyen, Thao %A Le, Sy Vinh %A Ho, Hai %A Le, Si Quang %D 2016 %F SisLab:2071 %I IEEE %J IEEE/ACM Transactions on Computational Biology and Bioinformatics %T Building ancestral recombination graphs for whole genomes %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2071/ %X Abstract— The Ancestral Recombination Graph (ARG) plays an important role in human population genetics. Nevertheless, most current ARG inference algorithms are only applicable to small data sets due to their computational burden. Margarita by Minichiello and Durbin [1] can handle larger data sets; however, it is still not feasible at genome scale. We hereby propose a heuristic algorithm, called ARG4WG, to construct plausible ARGs from thousands of whole chromosome samples, in which the so-called longest shared end, i.e. the longest match between left or right ends of sequences, is used for recombination in the building process. This strategy allows ARG4WG to significantly reduce the computational cost, by working hundreds to thousands times faster than Margarita. ARG4WG leads to ARGs with fewer numbers of recombination events on real data sets. Margarita is slightly better than ARG4WG in reconstructing tree topology on simulated data. The ARGs resulted from our algorithm also performed reasonably well in an association study with 5560 haplotypes across a whole Chromosome 11 of the Gambia dataset. These results indicate that ARG4WG is a good candidate for genome-wide association study from large data sets. %0 Journal Article %A Nguyen, The Hoang Anh %A Bui, Quyen %A Nguyen, Truong Thang %A Le, Thanh Ha %D 2019 %F SisLab:3745 %J Journal of Informatics and Mathematical Sciences %T A Robust PCA-SURE Thresholding Deep Neural Network Approach for Mental Task Brain Computer Interface %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3745/ %0 Journal Article %A Nguyen, The Hoang Anh %A Bui, The Duy %A Le, Thanh Ha %D 2020 %F SisLab:4176 %J Neural Computing and Applications %T A Deep Wavelet Sparse Autoencoder Method for Online and Automatic Electrooculographical Artifact Removal %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4176/ %0 Conference Paper %A Nguyen, The Hoang Anh %A Do, Anh Tuan %A Le, Thanh Ha %A Bui, The Duy %B Control, Automation and Systems (ICCAS), 2019 19th International Conference on %D 2019 %F SisLab:3743 %T A deep sparse autoencoder method for automatic EOG artifact removal %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3743/ %0 Conference Paper %A Nguyen, The Hoang Anh %A Le, Thanh Ha %A Bui, The Duy %B In Proceedings of 6th NAFOSTED Conference on Information and Computer Science (NICS), 2019 %D 2019 %F SisLab:3744 %T A Stacking Ensemble Learning Model for Mental State Recognition towards Implementation of Brain Computer Interface %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3744/ %0 Journal Article %@ ISSN: 2169-3536 %A Nguyen, The Hung %A Hoang, Xuan Huan %A Nguyen, P. Nam %A Nguyen, Vu Tam %A Dinh, Ngoc Thang %A NAM P. Nguyen, %A NGUYEN, TAM VU, %A THANG N. DINH, %D 2017 %F SisLab:2797 %J IEEE Access %P 15913-15926 %T Transitivity Demolition and the Fall of Social Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2797/ %V 5 %X In this paper, we study crucial elements of a complex network, namely its nodes and connections, which play a key role in maintaining the network’s structure and function under unexpected structural perturbations of nodes and edges removal. Specifically, we want to identify vital nodes and edges whose failure (either random or intentional) will break the most number of connected triples(or triangles)in the network. This problem is extremely important, because connected triples form the foundation of strong connections in many real-world systems, such as mutual relationships in social networks, reliable data transmission in communication networks, and stable routing strategies in mobile networks. Disconnected triples, analog to broken mutual connections, can greatly affect the network’s structure and disrupt its normal function, which can further lead to the corruption of the entire system. The analysis of such crucial elements will shed light on key factors behind the resilience and robustness of many complex systems in practice. We formulate the analysis under multiple optimization problems and show their intractability. We next propose efficient approximation algorithms, namely, DAK-n and DAK-e, which guarantee an (1 − 1/e)-approximate ratio (compared with the overall optimal solutions) while having the same time complexity as the best triangle counting and listing algorithm on power-lawnetworks.Thisadvantagemakes ouralgorithmsscaleextremelywellevenforverylargenetworks.Inanapplicationperspective,we perform comprehensive experiments on real social traces with millions of nodes and billions of edges. Empirical results indicate that our approaches achieve comparably better solution quality while are up to 100×faster than the current state-of-the-art methods. %0 Conference Paper %A Nguyen, The Huy %A Nguyen, Duc Thinh %A Nguyen, Nam Phong %A Phung, Nguyen Ngoc %A Truong, Anh Hoang %B SSC: the 2016 Student Scientific Contest %C Hanoi %D 2016 %F SisLab:1492 %T Meboo-the health & medical services management application %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1492/ %0 Conference Paper %A Nguyen, Thi Anh Dao %A Le, Trung Thanh %A Nguyen, Linh Trung %A Le, Vu Ha %B 2018 The 5th NAFOSTED Conference on Information and Computer Science (NICS) %C Ho Chi Minh city, Vietnam %D 2018 %F SisLab:3219 %T Nonnegative Tensor Decomposition for EEG Epileptic Spike Detection %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3219/ %0 Journal Article %@ 2588-1086 %A Nguyen, Thi Anh Dao %A Le, Trung Thanh %A Nguyen, Viet Dung %A Nguyen, Linh Trung %A Le, Vu Ha %D 2019 %F SisLab:3696 %I VNU %J VNU Journal of Science: Computer Science and Communication Engineering %T New feature selection method for multi-channel EEG epileptic spike detection system %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3696/ %X Epilepsy is one of the most common and severe brain disorders. Electroencephalogram (EEG) is widely used in epilepsy diagnosis and treatment, with it the epileptic spikes can be observed. Tensor decomposition-based feature extraction has been proposed to facilitate automatic detection of EEG epileptic spikes. However, tensor decomposition may still result in a large number of features which are considered negligible in determining expected output performance. We proposed a new feature selection method that combines the Fisher score and p-value feature selection methods to rank the features by using the longest common sequences (LCS) to separate epileptic and non-epileptic spikes. The proposed method significantly outperformed several state-of-the-art feature selection methods. %0 Conference Paper %A Nguyen, Thi Cham %A Pham, Thi Ngan %A Nguyen, Minh Chau %A Nguyen, Tri Thanh %A Ha, Quang Thuy %A Trường Đại học Công nghệ, ĐHQGHN, %A Trường Đại học Y Dược Hải Phòng, %B Asian Conference on Intelligent Information and Database Systems %C Phuket, Thailand %D 2020 %F SisLab:4043 %P 575-585 %T A Lifelong Sentiment Classification Framework Based on a Close Domain Lifelong Topic Modeling Method %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4043/ %V 12033 %X In lifelong machine learning, the determination of the hypotheses related to the current task is very meaningful thanks to the reduction of the space to look for the knowledge patterns supporting for solving the current task. However, there are few studies for this problem. In this paper, we propose the definitions for measuring the “close domains to the current domain”, and a lifelong sentiment classification method based on using the close domains for topic modeling the current domain. Experimental results on sentiment datasets of product reviews from Amazon.com show the promising performance of system and the effectiveness of our approach. %0 Conference Paper %A Nguyen, Thi Hanh %A Dang, Duc Hanh %B The 2018 International Conference on Knowledge and Systems Engineering (KSE) %C Vietnam %D 2018 %F SisLab:3151 %P 264-269 %T An Approach for Testing Model Transformations %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3151/ %0 Conference Paper %A Nguyen, Thi Hanh %A Dang, Duc Hanh %A Nguyen, Quang Trung %B KSE: the 2019 International Conference on Knowledge and Systems Engineering %C Vietnam %D 2019 %F SisLab:3578 %T On Analyzing Rule-Dependencies to Generate Test Cases for Model Transformations %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3578/ %X Quality model transformations play a key role in the successful realization of Model Driven Engineering in practice. In the relational model transformations, rule dependency relations directly impact quality properties such as correctness, completeness, and information preservation. The analysis of rule dependencies from the declarative specification is expected to bring advantages for testing transformation properties. In this paper, we proposed a black-box approach for testing relational model transformations based on the analysis of the declarative specification using Triple Graph Grammar (TGG) rules. We exploit declarative TGG rules to capture the rule dependencies. Then, rule dependencies are combined together using the t-way testing technique to create test case descriptions. We transform patterns representing the input test condition and the oracle function of a test case description into OCL (Object Constraint Language) constraints to facilitate automatically generating input test models by solving constraints and querying interesting properties on the output models. %0 Journal Article %@ 1932-6203 %A Nguyen, Thi Hau %A Berry, Vincent %A Ranwez, Vincent %A Scornavacca, Celine %D 2013 %F SisLab:1861 %J PLOS ONE %T Support Measures to Estimate the Reliability of Evolutionary Events Predicted by Reconciliation Methods %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1861/ %X The genome content of extant species is derived from that of ancestral genomes, distorted by evolutionary events such as gene duplications, transfers and losses. Reconciliation methods aim at recovering such events and at localizing them in the species history, by comparing gene family trees to species trees. These methods play an important role in studying genome evolution as well as in inferring orthology relationships. A major issue with reconciliation methods is that the reliability of predicted evolutionary events may be questioned for various reasons: Firstly, there may be multiple equally optimal reconciliations for a given species tree–gene tree pair. Secondly, reconciliation methods can be misled by inaccurate gene or species trees. Thirdly, predicted events may fluctuate with method parameters such as the cost or rate of elementary events. For all of these reasons, confidence values for predicted evolutionary events are sorely needed. It was recently suggested that the frequency of each event in the set of all optimal reconciliations could be used as a support measure. We put this proposition to the test here and also consider a variant where the support measure is obtained by additionally accounting for suboptimal reconciliations. Experiments on simulated data show the relevance of event supports computed by both methods, while resorting to suboptimal sampling was shown to be more effective. Unfortunately, we also show that, unlike the majority-rule consensus tree for phylogenies, there is no guarantee that a single reconciliation can contain all events having above 50% support. In this paper, we detail how to rely on the reconciliation graph to efficiently identify the median reconciliation. Such median reconciliation can be found in polynomial time within the potentially exponential set of most parsimonious reconciliations. %0 Journal Article %@ 0302-9743 %A Nguyen, Thi Hau %A Doyon, Jean-Philippe %A Pointet, Stéphanie %A Arigon Chifolleau, Anne-Muriel %A Ranwez, Vincent %A Berry, Vincent %D 2012 %F SisLab:1858 %J Lecture Notes in Computer Science %P 123-134 %T Accounting for Gene Tree Uncertainties Improves Gene Trees and Reconciliation Inference %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1858/ %V 7534 %X We propose a reconciliation heuristic accounting for gene duplications, losses and horizontal transfers that specifically takes into account the uncertainties in the gene tree. Rearrangements are tried for gene tree edges that are weakly supported, and are accepted whenever they improve the reconciliation cost. We prove useful properties on the dynamic programming matrix used to compute reconciliations, which allows to speed-up the tree space exploration when rearrangements are generated by Nearest Neighbor Interchanges (NNI) edit operations. Experimental results on simulated and real data confirm that running times are greatly reduced when considering the above-mentioned optimization in comparison to the naïve rearrangement procedure. Results also show that gene trees modified by such NNI rearrangements are closer to the correct (simulated) trees and lead to more correct event predictions on average. The program is available at http://www.atgc-montpellier.fr/ Mowgli/ %0 Journal Article %A Nguyen, Thi Hau %A Lu, Dang Nhac %A Nguyen, Duc Nhan %A Nguyen, Ha Nam %D 2020 %F SisLab:3928 %I MDPI %J Electronics %N 2 %T Dynamic Basic Activity Sequence Matching Method in Abnormal Driving Pattern Detection Using Smartphone Sensors %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3928/ %V 9 %X In this work, we present a novel method, namely dynamic basic activity sequence matching (DAS), a combination of machine learning methods and flexible threshold based methods for distinguishing normal and abnormal driving patterns. Indeed, DAS relies on the activity detection module (ADM) presented in our previous work to analyze each driving pattern as a sequence of basic activities—stopping (S), going straight (G), turning left (L), and turning right (R). In fact, the threshold value and other parameters like the duration of long and short activities are iteratively induced from the collected dataset. Hence, DAS is flexible and independent of driving contexts such as vehicle modes and road conditions. Experimental results, on the dataset collected from numerous motorcyclists, show the outperformance of our proposed method against dynamic time warping and the two popular machine learning methods—random forest and neural network—in distinguishing the normal and abnormal driving patterns. Moreover, we propose an efficient framework composing of two phases: in the first phase, the normal and abnormal driving patterns are distinguished by relying on DAS. In the second phase, the detected abnormal patterns are further classified into various specific abnormal driving patterns—weaving, sudden braking, etc. This fusion framework again achieves the highest overall accuracy of 97.94%. %0 Journal Article %@ 1748-7188 %A Nguyen, Thi Hau %A Ranwez, Vincent %A Pointet, Stéphanie %A Chifolleau, Anne-Muriel %A Doyon, Jean-Philippe %A Berry, Vincent %D 2013 %F SisLab:1859 %J Algorithms for Molecular Biology %N 1 %P 12 %T Reconciliation and local gene tree rearrangement can be of mutual profit %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1859/ %V 8 %X Reconciliation methods compare gene trees and species trees to recover evolutionary events such as duplications, transfers and losses explaining the history and composition of genomes. It is well-known that gene trees inferred from molecular sequences can be partly erroneous due to incorrect sequence alignments as well as phylogenetic reconstruction artifacts such as long branch attraction. In practice, this leads reconciliation methods to overestimate the number of evolutionary events. Several methods have been proposed to circumvent this problem, by collapsing the unsupported edges and then resolving the obtained multifurcating nodes, or by directly rearranging the binary gene trees. Yet these methods have been defined for models of evolution accounting only for duplications and losses, i.e. can not be applied to handle prokaryotic gene families. Results We propose a reconciliation method accounting for gene duplications, losses and horizontal transfers, that specifically takes into account the uncertainties in gene trees by rearranging their weakly supported edges. Rearrangements are performed on edges having a low confidence value, and are accepted whenever they improve the reconciliation cost. We prove useful properties on the dynamic programming matrix used to compute reconciliations, which allows to speed-up the tree space exploration when rearrangements are generated by Nearest Neighbor Interchanges (NNI) edit operations. Experiments on synthetic data show that gene trees modified by such NNI rearrangements are closer to the correct simulated trees and lead to better event predictions on average. Experiments on real data demonstrate that the proposed method leads to a decrease in the reconciliation cost and the number of inferred events. Finally on a dataset of 30 k gene families, this reconciliation method shows a ranking of prokaryotic phyla by transfer rates identical to that proposed by a different approach dedicated to transfer detection [BMCBIOINF 11:324, 2010, PNAS 109(13):4962–4967, 2012]. Conclusions Prokaryotic gene trees can now be reconciled with their species phylogeny while accounting for the uncertainty of the gene tree. More accurate and more precise reconciliations are obtained with respect to previous parsimony algorithms not accounting for such uncertainties [LNCS 6398:93–108, 2010, BIOINF 28(12): i283–i291, 2012]. A software implementing the method is freely available at http://www.atgc-montpellier.fr/Mowgli/. %0 Conference Paper %A Nguyen, Thi Hoai Thu %A Le, Trung Thanh %A Chu, Thi Phuong Dung %A Nguyen, Linh Trung %A Le, Vu Ha %B 2017 International Conference on Advanced Technologies for Communications (ATC) %C Quy Nhon, Viet Nam %D 2017 %F SisLab:2596 %T Multi-source Data Analysis for Bike Sharing Systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2596/ %X Bike sharing systems (BSSs) have become common in many cities worldwide, providing a new transportation mode for residents’ commutes. However, the management of these systems gives rise to many problems. As the bike pick-up demands at different places are unbalanced at times, the systems have to be rebalanced frequently. Rebalancing the bike availability effectively, however, is very challenging as it demands accurate prediction for inventory target level determination. In this work, we propose two types of regression models using multi-source data to predict the hourly bike pick-up demand at cluster level: Similarity Weighted K-Nearest-Neighbor (SWK) based regression and Artificial Neural Network (ANN). SWK-based regression models learn the weights of several meteorological factors and/or taxi usage and use the correlation between consecutive time slots to predict the bike pick-up demand. The ANN is trained by using historical trip records of BSS, meteorological data, and taxi trip records. Our proposed methods are tested with real data from a New York City BSS: Citi Bike NYC. Performance comparison between SWK-based and ANN-based methods is provided. Experimental results indicate the high accuracy of ANN-based prediction for bike pick-up demand using multi-source data. %0 Conference Paper %A Nguyen, Thi Hong Khanh %A Ha, Quang Thuy %A VNU-University of Eng. and Tech, %A Electric Power University, %B 2018 15th International Conference on Control, Automation, Robotics and Vision (ICARCV) %C Marina Bay Sands Expo and Convention Centre, Singapore %D 2018 %F SisLab:3300 %T A Learning Method based on Bisimulation in the Inconsistent Knowledge Systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3300/ %X Inconsistencies may naturally occur in the considered application domains in Artificial Intelligence, for example as a result of data mining works in distributed sources. In order to solve inconsistent knowledge, several paraconsistent description logics have been proposed. In this paper, we face the problem of concept learning for an inconsistent knowledge base system based on bisimulation. This algorithm allows learning a concept from a training information system in a paraconsistent descriptive logic system with a set of positive items, negative items, and inconsistent items. Here, we present a system for learning concept in an inconsistent knowledge base and discuss preliminary experimental results obtained in the electronic application domain. %0 Conference Paper %A Nguyen, Thi Hong Khanh %A Ha, Quang Thuy %A Nguyen, Anh Linh %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1536 %T Logic-used Methods Solving Uncertainty and Inconsistency in Knowledge Intergration and Applications %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1536/ %X Working out representation methods for inconsistent knowledge is one of the most important tasks in the Knowledge Integration field. It is the representation structures, which decide the way how a knowledge system collects and processes the conflict knowledge, and thus it influences directly the ability and effectiveness of the system. One effective way to present inconsistent knowledge is using paraconsistent logics. In this paper, I present a case study of several paraconsistent logics that allow representing inconsistency and processing it. In each of these logics I will present the semantic, the proof theory and the applicability. With the aim to develop the representation of inconsistent information in a distributed system, I propose a model of paraconsistent logic for distributed system. %0 Conference Paper %A Nguyen, Thi Hong Khanh %A Tran, Trong Hieu %A Ha, Quang Thuy %B 15th International Conference on Control, Automation, Robotics and Vision, ICARCV 2018 %C Singapore %D 2018 %F SisLab:3436 %P 430-435 %T A Learning Method based on Bisimulation in Inconsistent Knowledge Systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3436/ %X Inconsistencies may naturally occur in the considered application domains in Artificial Intelligence, for example as a result of data mining works in distributed sources. In order to solve inconsistent knowledge, several paraconsistent description logics have been proposed. In this paper, we face the problem of learning concept for inconsistent knowledge base system based on bisimulation. This algorithm allows learning a concept from a training information system in paraconsistent descriptive logic system with a set of positive items, negative items and inconsistent items. Here, we present a system for learning concept in inconsistent knowledge base and discuss preliminary experimental results obtained in the electronic application domain. %0 Conference Paper %A Nguyen, Thi Huong Thao %A Vu, Huu Tien %A Vu, Van San %A Hoang, Van Xiem %A Le, Thanh Ha %A Dinh, Trieu Duong %B the 2016 International Conference Advanced Technologies for Communications (ATC) %C Hanoi, Vietnam %D 2016 %F SisLab:2036 %T Side information creation using adaptive block size for distributed video coding %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2036/ %0 Conference Paper %A Nguyen, Thi Huong Thuy %A Hoang, Xuan-Huan %A Nguyen, Ngoc Ky %B 2013 International Conference on Computing, Management and Telecommunications (ComManTel) %C HoChiMinh city %D 2013 %F SisLab:135 %K Correlation;Databases;Deformable models;Fingerprint recognition;Fingers;Image matching;Nonlinear distortion;Elastic deformations;Fingerprint verification;Local deformation models;Minutiae matching;Thin-Plate-Spline models %P 334-339 %T An Efficient Method for Fingerprint Matching based on Local Point Model %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/135/ %X This paper proposes a fingerprint matching method based on local Thin-Plate-Spline (TPS) deformation model, a warping technique, to deal with non-linear distorted fingerprints. After determining the set of corresponding minutiae pairs between two fingerprints by using an affine transformation, a set of corresponding pseudo-minutiae pairs are created based on their local ridge-valley structure comparison. These points are associated with the known corresponding point pairs to select a suitable landmark point set for using Local Thin Plate Spline deformation model over 9 partial areas of fingerprint images in order to find new corresponding minutiae pairs. This procedure is repeated until no more new corresponding minutiae pairs are distinguished or the number of corresponding point pairs is large enough. The experimental results on the database FVC2004 show that the proposed method significantly improves matching performance compared to the global TPS warping method. %0 Journal Article %@ ‎0018-9340 %A Nguyen, Thi Huyen Chau %A Richard, Pascal %A Grolleau, Emmanuel %D 2015 %F SisLab:1474 %I IEEE %J IEEE Transactions on Computers %N 7 %P 1805-1818 %T An FTPAS for Response Time Analysis of Fixed Priority Real-Time Tasks with Resource Augmentation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1474/ %V 64 %0 Conference Paper %A Nguyen, Thi Huyen Chau %A Tran, N.S. %A Le, V.H. %A Richard, Pascal %B 2013: the 21st International conference on Real-Time Networks and Systems %D 2013 %F SisLab:774 %P 265-274 %T Approximation Scheme for Real-Time Tasks under Fixed-Priority Scheduling with Deferred Preemption %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/774/ %0 Conference Paper %A Nguyen, Thi Minh Trang %A Tran, Van Lien %A Can, Duy Cat %A Ha, Quang Thuy %A Vu, Thi Ly %A Chng, Eng Siong %A Faculty of Information Technology (FIT), %B ICMLSC 2019: the 3rd International Conference on Machine Learning and Soft Computing %C Da Lat, Viet Nam %D 2019 %F SisLab:3474 %P 221-225 %T QASA: Advanced Document Retriever for Open-Domain Question Answering by Learning to Rank Question-Aware Self-Attentive Document Representations %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3474/ %X For information consumers, being able to obtain a short and accurate answer for a query is one of the most desirable features. This motivation, along with the rise of deep learning, has led to a boom in open-domain Question Answering (QA) research. While the problem of machine comprehension has received multiple success with the help of large training corpora and the emergence of attention mechanism, the development of document retrieval in open-domain QA is lagged behind. In this work, we propose a novel encoding method for learning question-aware self-attentive document representations. By applying pair-wise ranking approach to these encodings, we build a Document Retriever, called QASA, which is then integrated with a machine reader to form a complete open-domain QA system. Our system is thoroughly evaluated using QUASAR-T dataset and shows surpassing results compared to other state-of-the-art methods. %0 Conference Paper %A Nguyen, Thi Ngoc Diep %B the IEEE Conference on Computer Vision and Pattern Recognition Workshops, Real-World Recognition from Low-Quality images and Video (RLQ@ICCV’19) %C Seoul, Republic of Korea %D 2019 %F SisLab:3791 %T State-of-the-Art in Action: Unconstrained Text Detection %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3791/ %X In this paper, we stage five real-world scenarios for six state-of-the-art text detection methods in order to evaluate how competent they are with new data without any training process. Moreover, this paper analyzes the architecture design of those methods to reveal the influence of pipeline choices on the detection quality. The setup of experimental studies are straight-forward: we collect and manually annotate test data, we reimplement the pretrained models of the state-of-the-art methods, then we evaluate and analyze how well each method achieve in each of our collected datasets. We found that most of the state-of-the-art methods are competent at detecting textual information in unseen data, however, some are more readily used for real-world applications. Surprisingly, we also found that the choice of a post-processing algorithm correlates strongly with the performance of the corresponding method. We expect this paper would serve as a reference for researchers as well as application developers in the field. All collected data with ground truth annotation and their detected results is publicly available at our Github repository: https://github.com/chupibk/ HBlab-rlq19. %0 Conference Paper %A Nguyen, Thi Nhan %A Hoang, Xuan Tung %A Bui, Ngoc Dung %B Hội nghị Khoa học công nghệ lần thứ XXII - Trường Đại học Giao Thông Vận Tải %C Ha noi %D 2020 %F SisLab:4279 %P 276-280 %T MÔ HÌNH HỆ THỐNG QUẢN TRỊ TIN NHẮN DỰA TRÊN MODULE MEI-CHAT CHO CÁC TRANG WEB CÓ SỐ LƯỢNG NGƯỜI DÙNG LỚN %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4279/ %X Dịch vụ chăm sóc khách hàng trực tuyến trên các trang web đang là một thành phần quan trọng để duy trì hoạt động kinh doanh trên môi trường mạng. Các hộp thoại nhắn tin trực tuyến được nhúng vào các trang web cung cấp khả năng liên lạc giữa người truy cập và quản trị viên trang web. Trong bài báo này này, chúng tôi giới thiệu module mei-chat xây dựng bằng các công nghệ Web tiên tiến nhất như NodeJS, AngularJS, SocketIO. Mei-chat gồm ba thành chính là meiChatCustomer, meiChatAdmin và meiChatBackend. MeiChatBackend được xây dựng bằng Nodejs tích hợp SocketIO với cấu trúc cân bằng tải nhằm đảm bảo số lượng lớn người kết nối. MeiChatAdmin cho phép quản trị viên nhắn tin, xem lịch sử tin nhắn của người truy cập trang web. MeiChatCustomer là hộp thoại tin nhắn có thể dễ dàng tích hợp vào các trang web, cho phép người truy cập nhắn tin với quản trị viên. %0 Conference Paper %A Nguyen, Thi Nhat Thanh %A Bui, Quang Hung %A Luong, Chinh Ke %A Luu, Viet Hung %A Pham, Van Ha %A Dao, Ngoc Thanh %A Pham, Huu Bang %A Man, Duc Chuc %A Le, Thanh Ha %A Nguyen, Nam Hoang %A Nguyen, Hai Chau %A Nguyen, Thanh Thuy %B BESETOHA %C Ha Noi, Viet Nam %D 2014 %F SisLab:1420 %T Air Pollution Monitoring and Warning System %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1420/ %X Nowaday, remote sensing images have provided a large dataset with geospatial information at global scale at different resolutions, which is widely using in various domains. The usage of satellite technology for air pollution monitoring applications has been recently increasing especially to provide global-to-local distribution of aerosol and its properties for deriving Particulate Matter concentration (PM). The paper investigates aerosol retrieval for multi- resolution satellite images. After that, PM is estimated from aerosol products and meteorological parameters in order to provide dust observations at different spatial scales. A geographic information system for air pollution monitoring and warning is developed based on these research results. %0 Conference Paper %A Nguyen, Thi Nhat Thanh %A Bui, Quang Hung %A Luong, Chinh Ke %A Luu, Viet Hung %A Pham, Van Ha %A Dao, Ngoc Thanh %A Pham, Huu Bang %A Man, Duc Chuc %A Le, Thanh Ha %A Nguyen, Nam Hoang %A Nguyen, Hai Chau %A Nguyen, Thanh Thuy %B 2014: Hội thảo ứng dụng GIS toàn quốc 2014 %D 2014 %F SisLab:1422 %P 863-871 %T Nghiên cứu và phát triển hệ thống giám sát và cảnh báo mức độ ô nhiễm không khí tại Việt Nam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1422/ %V 2 %0 Journal Article %@ ‎1748-9326 %A Nguyen, Thi Nhat Thanh %A Bui, Quang Hung %A Pham, Van Hung %A Luu, Viet Hung %A Chu, Duc Man %D 2015 %F SisLab:1609 %J Environmental Research Letters %T Particulate matter concentration mapping from MODIS satellite data: a Vietnamese case study %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1609/ %V 9-10 %X Particulate Matter (PM) pollution is one of the most important air quality concerns in Vietnam. In this study, we integrate ground-based measurements, meteorological and satellite data to map temporal PM concentrations at a 10 × 10 km grid for the entire of Vietnam. We specifically used MODIS Aqua and Terra data and developed statistically-significant regression models to map and extend the ground-based PM concentrations. We validated our models over diverse geographic provinces i.e., North East, Red River Delta, North Central Coast and South Central Coast in Vietnam. Validation suggested good results for satellite-derived PM2.5 data compared to ground-based PM2.5 (n = 285, r2 = 0.411, RMSE = 20.299 μg m−3 and RE = 39.789%). Further, validation of satellite-derived PM2.5 on two independent datasets for North East and South Central Coast suggested similar results (n = 40, r2 = 0.455, RMSE = 21.512 μg m−3, RE = 45.236% and n = 45, r2 = 0.444, RMSE = 8.551 μg m−3, RE = 46.446% respectively). Also, our satellite-derived PM2.5 maps were able to replicate seasonal and spatial trends of ground-based measurements in four different regions. Our results highlight the potential use of MODIS datasets for PM estimation at a regional scale in Vietnam. However, model limitation in capturing maximal or minimal PM2.5 peaks needs further investigations on ground data, atmospheric conditions and physical aspects. %0 Conference Paper %A Nguyen, Thi Nhat Thanh %A Ha, Duc Van %A Do, Thi Nhu Ngoc %A Nguyen, Van Hai %A Ngo, Xuan Truong %A Pham, Van Ha %A Nguyen, Ngoc Duc %A Bui, Quang Hung %B 4th International Forum on Sustainable Future in Asia / 4th NIES International Forum %C Pan Pacific Hanoi, Vietnam %D 2019 %F SisLab:3499 %T Air pollution monitoring network using low-cost sensors, a case study in Hanoi, Vietnam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3499/ %X Air pollution is a serious problem in Vietnam, especially in urban areas with high pressures of population, traffic, construction, and industrial development. Besides high accurate measurements from automatic and continuous monitoring ground stations and high-cost sensor devices, low-cost sensors have recently utilized to extent air pollution monitoring networks although their data quality is still argumentative. In this paper, we present a low-cost device, named FAirKit, which measured 6 basic air pollutants including PM2.5, PM10, CO, O3, NO2, and SO2, and temperature and relative humidity. The sensors are calibrated with standard devices to improve their data quality. FAirKits are installed and transferred data in real-time to servers where an information system based on Sensor Web Enablement (SWE) standard of Open Geospatial Consortium (OGC) has been developed to store, process, and visualize real-time air pollution information. Currently, the low-cost sensors network has been deploying in Hanoi, Vietnam to enhance public awareness and alert local people to air pollution. %0 Journal Article %@ 1343-8808 %A Nguyen, Thi Nhat Thanh %A Le, Hoang Anh %A Mac, Thi Minh Tra %A Nguyen, Thi Trang Nhung %A Pham, Van Ha %A Bui, Quang Hung %D 2018 %F SisLab:3129 %I Association of International Research Initiatives for Environmental Studies (AIRIES), Japan %J Global Environmental Research %N 1&2 %T Current Status of PM2.5 Pollution and its Mitigation in Vietnam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3129/ %V 22 %X Vietnam is experiencing serious fine particulate matter (PM2.5) pollution as a result of local activities and long-range transport (LRT) pollutants. In this article, we summarize and analyze PM2.5 data from ground stations and manual measurements showing PM2.5 status, characteristics and emission sources in the period from 1996 to 2017 in Vietnam. In addition, we provide a brief impact assessment of PM2.5 pollution on public health regarding diseases and deaths. Conscious of PM2.5's harmful effects, Vietnam has been taking steps to mitigate PM2.5 pollution in various forms through efforts by the government, non-governmental organizations, media, communities and individuals, and has obtained initial results. This article presents a comprehensive review of current PM2.5 pollution and its mitigation in Vietnam. %0 Book Section %A Nguyen, Thi Nhat Thanh %A Luu, Viet Hung %A Pham, Van Ha %A Bui, Quang Hung %A Nguyen, Thi Kim Oanh %B TORUS 3 – Toward an Open Resource Using Services %D 2020 %F SisLab:4320 %I ISTE Ltd 2020. Published by ISTE Ltd and John Wiley & Sons, Inc %P 103-130 %T Particulate Matter Concentration Mapping from Satellite Imagery %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4320/ %0 Journal Article %@ 0269-7491 %A Nguyen, Thi Nhat Thanh %A Pham, Ha V. %A Lasko, Kristofer %A Bui, Mai T. %A Laffly, Dominique %A Jourdan, Astrid %A Bui, Quang Hung %D 2019 %F SisLab:3798 %J Environmental Pollution %P 113106 %T Spatiotemporal analysis of ground and satellite-based aerosol for air quality assessment in the Southeast Asia region %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3798/ %V 255 %X Satellite observations for regional air quality assessment rely on comprehensive spatial coverage, and daily monitoring with reliable, cloud-free data quality. We investigated spatiotemporal variation and data quality of two global satellite Aerosol Optical Depth (AOD) products derived from MODIS and VIIRS imagery. AOD is considered an essential atmospheric parameter strongly related to ground Particulate Matter (PM) in Southeast Asia (SEA). We analyze seasonal variation, urban/rural area influence, and biomass burning effects on atmospheric pollution. Validation indicated a strong relationship between AERONET ground AOD and both MODIS AOD (R2 = 0.81) and VIIRS AOD (R2 = 0.68). The monthly variation of satellite AOD and AERONET AOD reflects two seasonal trends of air quality separately for mainland countries including Myanmar, Laos, Cambodia, Thailand, Vietnam, and Taiwan, Hong Kong, and for maritime countries consisting of Indonesia, Philippines, Malaysia, Brunei, Singapore, and Timor Leste. The mainland SEA has a pattern of monthly AOD variation in which AODs peak in March/April, decreasing during wet season from May–September, and increasing to the second peak in October. However, in maritime SEA, AOD concentration peaks in October. The three countries with the highest annual satellite AODs are Singapore, Hong Kong, and Vietnam. High urban population proportions in Singapore (40.7%) and Hong Kong (21.6%) were associated with high AOD concentrations as expected. AOD values in SEA urban areas were a factor of 1.4 higher than in rural areas, with respective averages of 0.477 and 0.336. The AOD values varied proportionately to the frequency of biomass burning in which both active fires and AOD peak in March/April and September/October. Peak AOD in September/October in some countries could be related to pollutant transport of Indonesia forest fires. This study analyzed satellite aerosol product quality in relation to AERONET in SEA countries and highlighted framework of air quality assessment over a large, complicated region. %0 Conference Paper %A Nguyen, Thi Nhat Thanh %A Ta, Viet Cuong %A Le, Thanh Ha %A Mantovani, Simone %B 2013: The Fifth International Conference on Knowledge and Systems Engineering (KSE 2013) %C Hanoi, Vietnam %D 2013 %F SisLab:788 %P 351-362 %T Particulate Matter Concentration Estimation from Satellite Aerosol and Meteorological Parametters: Data-Driven Aproaches %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/788/ %V 1 %0 Conference Paper %A Nguyen, Thi Nhat Thanh %A Ta, Viet Cuong %A Le, Thanh Ha %A Mantovani, Simone %B KSE: the 6th International Conference on Knowledge and Systems Engineering %C Hanoi, Vietnam %D 2014 %F SisLab:1911 %I Springer %P 351-362 %T Particulate matter concentration estimation from satellite aerosol and meteorological parameters: data-driven approaches %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1911/ %0 Conference Paper %A Nguyen, Thi Nhung %A Perry, Stuart %A Le, Thanh Ha %A Nguyen, Thi Thuy %B The 9th International Symposium On Information And Communication Technology (SoICT2018) %D 2018 %F SisLab:3291 %T Pavement Crack Detection using Convolutional Neural Network %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3291/ %0 Conference Paper %A Nguyen, Thi Phuong Thao %A Le, Sy Vinh %B 2017 4th NAFOSTED Conference on Information and Computer Science (NICS 2017) %C Hanoi, Vietnam %D 2017 %F SisLab:2684 %P 248-253 %T Building minimum recombination ancestral recombination graphs for whole genomes %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2684/ %X Building ancestral recombination graphs (ARGs) with the minimum number of recombination events is an essential problem in genomic population studies. Exhaustive search methods have been proposed to solve the problem with few sequences and markers. Recently, we introduced ARG4WG algorithm [1] to build plausible ARGs for whole genomes of thousand samples. However, the ARG4WG algorithm was not designed to optimize the number of recombination events. In this work, we propose a new design of the ARG4WG algorithm to optimize the number of recombination events. The key idea is to analyze and combine different criteria (i.e., the longest shared ends, the maximum similarity between sequences, and the length of sequences) to select the best candidates for recombinations. The experiments on different data sets showed that our new algorithms produced ARGs with smaller number of recombination events in comparison to other methods. The methods allow us to build minimum recombination ARGs from whole genome data sets. %0 Conference Paper %A Nguyen, Thi Phuong Thao %A Le, Sy Vinh %A Ho, Bich Hai %A Le, Si Quang %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1537 %T Building ancestral recombination graphs for whole genomes %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1537/ %X Ancestral Recombination Graph (ARG) plays an important role in human population genetics. Nevertheless, most of current ARG inference algorithms are only applicable to small data sets due to their computational burden. Margarita by Minichiello and Durbin [1] can handle larger data sets; however, it is still not feasible at genome scale. We hereby propose a heuristic algorithm, called ARG4WG, to construct plausible ARGs from thousands of whole chromosome samples, in which the so-called longest shared end, i.e. the longest match between left or right ends of sequences, is used for recombination in the building process. This strategy not only allows ARG4WG to significantly reduce the computational cost, by hundreds to thousands times faster than Margarita but also leads to ARGs with fewer number of recombination events. The ARGs resulted from our algorithm also perform reasonably well in association study with 5560 haplotypes across whole chromosome 11 of Gambia dataset. These results indicate that ARG4WG is a good candidate for genome-wide association study from large data sets. %0 Conference Paper %A Nguyen, Thi Thu Trang %A Nguyen, Dai Tho %A Vu, Duy Loi %A VNU University of Engineering and Technology, %B The 15th IEEE-RIVF International Conference on Computing and Communication Technologies (RIVF 2021) %C Hanoi, Vietnam %D 2021 %F SisLab:4445 %K Malware classification, machine learning, k-nearest neighbors algorithms, prototype-based learning, hypercuboids %T A Hypercuboid-Based Machine Learning Algorithm for Malware Classification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4445/ %X Malware attacks have been among the most serious threats to cyber security in the last decade. Anti-malware software can help safeguard information systems and minimize their exposure to the malware. Most of anti-malware programs detect malware instances based on signature or pattern matching. Data mining and machine learning techniques can be used to automatically detect models and patterns behind different types of malware variants. However, traditional machine-based learning techniques such as SVM, decision trees and naive Bayes seem to be only suitable for detecting malicious code, not effective enough for complex problems such as classification. In this article, we propose a new prototype extraction method for non-traditional prototype-based machine learning classification. The prototypes are extracted using hypercuboids. Each hypercuboid covers all training data points of a malware family. Then we choose the data points nearest to the hyperplanes as the prototypes. Malware samples will be classified based on the distances to the prototypes. Experiments results show that our proposition leads to F1 score of 96.5% for classification of known malware and 97.7% for classification of unknown malware, both better than the original prototype-based classification method. %0 Journal Article %@ 2525-2224 %A Nguyen, Thi Thu Trang %A Nguyen, Dai Tho %A Vu, Duy Loi %A Trường Đại học Công nghệ, Đại học Quốc gia Hà Nội, %D 2019 %F SisLab:3755 %I Posts and Telecommunications Institute of Technology %J Journal of Science and Technology on Information and Communications %T Phân tích mã độc dựa trên các kỹ thuật n-gram và học máy %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3755/ %X Mã độc đang là mối đe dọa lớn đến an ninh của các hệ thống máy tính. Vì vậy phân loại mã độc để có những biện pháp đối phó thích hợp là một phần quan trọng trong lĩnh vực an toàn thông tin. Trong bài báo này chúng tôi cải tiến giải thuật trích rút điểm mẫu trong quy trình phân loại mở mã độc dựa trên điểm mẫu được đề xuất bởi Rieck và các công sự [1]. Chúng tôi áp dụng tư tưởng của hai phương pháp phân loại dựa trên điểm đặc trưng là Dendroid [2] và phương pháp được đề xuất bởi P. Shrestha và các cộng sự [3], trích rút các điểm mẫu trên từng họ mã độc thay vì dựa trên toàn bộ các họ mã độc. Thực nghiệm cho thấy phương pháp của chúng tôi cho kết quả phân loại tốt với khả năng nhận biết mã độc là 0.981% và phát hiện mã độc mới là 0.988% cao hơn phiên bản gốc của phương pháp dựa trên điểm mẫu [1]. %0 Conference Paper %A Nguyen, Thi Thu Trang %A Vuong, Thi Hai Yen %A Tran, Van Lien %A Nguyen, Le Minh %A Phan, Xuan Hieu %B The 12th IEEE International Conference on Knowledge and Systems Engineering (KSE) %C Can Tho, Vietnam %D 2020 %F SisLab:4218 %T Key phrase Generation for Vietnamese Administrative Documents: A Collaborative Approach %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4218/ %0 Conference Paper %A Nguyen, Thi Xuan Huong %A Dang, Tran Thai %A Le, Anh Cuong %B 2014: The Third Asian Conference on Information Systems (ACIS 2014) %C Nha Trang, Viet Nam %D 2014 %F SisLab:1181 %T Adapting Vietnamese Word Segmentation for Microblogs-Style Data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1181/ %0 Conference Paper %A Nguyen, Thi Xuan Huong %A Nguyen, The Tung %A Le, Anh Cuong %A Dang, Tran Thai %B 2014: The 6th International Conference on Knowledge and Systems Engineering (KSE 2014) %C Hanoi, Vietnam %D 2014 %F SisLab:447 %P 617-627 %T Using Large N-gram for Vietnamese Spell Checking %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/447/ %0 Conference Paper %A Nguyen, Thi-Hanh %A Dang, Duc Hanh %B The 8th NAFOSTED Conference on Information and Computer Science (NICS 2021) %D 2021 %F SisLab:4667 %P 1-6 %T A Graph Analysis Based Approach for Specification-Driven Testing of Model Transformations %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4667/ %0 Journal Article %@ 2588-1086 %A Nguyen, Thi-Hau %A Do, Trung-Tuan %A Nguyen, Duc-Nhan %A Lu, Dang Nhac %A Nguyen, Ha Nam %D 2020 %F SisLab:3978 %I VNU %J VNU Journal of Science: Computer Science and Communication Engineering %N 1 %T A Hybrid Method Based on Genetic Algorithm and Ant Colony System for Traffic Routing Optimization %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3978/ %V 36 %X This paper presents a hybrid method that combines the genetic algorithm (GA) and the ant colony system algorithm (ACS), namely GACS, to solve the traffic routing problem. In the proposed framework, we use the genetic algorithm to optimize the ACS parameters in order to attain the best trips and travelling time through several novel functions to help ants to update the global and local pheromones. The GACS framework is implemented using the VANETsim package and the real city maps from the open street map project. The experimental results show that our framework achieves a considerably higher performance than A-Star and the classical ACS algorithms in terms of the length of the global best path and the time for trips. Moreover, the GACS framework is also efficient in solving the congestion problem by online monitoring the conditions of traffic light systems. %0 Journal Article %@ 1573-7705 %A Nguyen, Thi-Thuy-Lien %A Pham, Tuan-Minh %A Pham, Manh Linh %D 2023 %F SisLab:4791 %I Springer Nature %J Journal of Network and Systems Management %N 18 %T Efficient Redundancy Allocation for Reliable Service Function Chains in Edge Computing %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4791/ %V 31 %X Ensuring the high reliability of service function chains (SFCs) in Edge Computing, in which several distributed edge servers are available, is a challenging issue. Previous studies on reliable SFCs ignore the impact of physical hardware failures when multiple virtual network functions (VNFs) are deployed on the same server, resulting in inaccurate reliability estimates. In this paper, we first propose an optimization model and approximation algorithm, considering both hardware and software reliability, to maximize the reliability of SFCs in each service demand. We then develop an algorithm to increase the reliability of SFCs to a given requirement. The evaluation results show that our algorithms achieve a near-optimal solution with a significant reduction in the computational time for finding the placement of redundant VNFs. We also observe that our proposed redundancy VNF allocation can efficiently save the backup cost to achieve a given SFC reliability requirement. %0 Journal Article %A Nguyen, Thu Trang %A Ngo, Kien Tuan %A Nguyen, Van Son %A Vo, Dinh Hieu %F SisLab:4619 %J IEEE Transactions on Software Engineering %T A Variability Fault Localization Approach for Software Product Lines %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4619/ %X Software fault localization is one of the most expensive, tedious, and time-consuming activities in program debugging. This activity becomes even much more challenging in Software Product Line (SPL) systems due to variability of failures. These unexpected behaviors are induced by variability faults which can only be exposed under some combinations of system features. The interaction among these features causes the failures of the system. Although localizing bugs in single-system engineering has been studied in-depth, variability fault localization in SPL systems still remains mostly unexplored. In this article, we present VarCop, a novel and effective variability fault localization approach. For an SPL system failed by variability bugs, VarCop isolates suspicious code statements by analyzing the overall test results of the sampled products and their source code. The isolated suspicious statements are the statements related to the interaction among the features which are necessary for the visibility of the bugs in the system. In VarCop, the suspiciousness of each isolated statement is assessed based on both the overall test results of the products containing the statement as well as the detailed results of the test cases executed by the statement in these products. On a large public dataset of buggy SPL systems, our empirical evaluation shows that VarCop significantly improves two state-of-the-art fault localization techniques by 33% and 50% in ranking the incorrect statements in the systems containing a single bug each. In about two-thirds of the cases, VarCop correctly ranks the buggy statements at the top-3 positions in the resulting lists. Moreover, for the cases containing multiple bugs, VarCop outperforms the state-of-the-art approaches 2 times and 10 times in the proportion of bugs localized at the top-1 positions. Especially, in 22% and 65% of the buggy versions, VarCop correctly ranks at least one bug in a system at the top-1 and top-5 positions. %0 Conference Paper %A Nguyen, Thu Trang %A Nguyen, Hue %A Bui, Quang Cuong %A Pham, Ngoc Hung %A Vo, Dinh Hieu %A Takeuchi, Shigeki %B 14th International Conference on Advanced COMPuting and Applications %D 2020 %F SisLab:4083 %P 6-8 %T Practical approach to access the impact of global variables on program parallelism %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4083/ %X Global variables may have a significant impact on preventing programs from automatic parallelism. This paper introduces a practical approach to measure the effect of global variables on program parallelism. First, we conduct static data dependence analysis among program variables and represent such dependencies by a Variable Dependence Graph. Then, we analyze this graph for measuring and identifying which global variables have a significant impact on program parallelism. To evaluate this approach, we conduct experiments on 20 benchmark programs and an industrial application. The experimental results show that half of the studied programs contain large impact variables which may be the cause of preventing programs from parallel execution. %0 Journal Article %A Nguyen, Thu Trang %A Truong, Ninh Thuan %A Nguyen, Viet Ha %A VNU-UET, %D 2011 %F SisLab:116 %J International Journal of Software Engineering and Knowledge Engineering %N 04 %P 605-619 %T Verifying Java Object Invariants at Runtime %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/116/ %V 21 %X An object invariant consisting of a set of properties that must hold for all instances of a class at any time is usually used in object-oriented design. However, verifying object invariants at runtime is always a challenging task in software verification. This paper proposes a method for verifying invariants of Java objects at runtime using AOP. Suppose that a software application is designed using UML models and its constraints are specified in OCL expressions, the software is then implemented, by default, using the UML design. They propose to construct verifiable aspects which are automatically generated from OCL constraints. These aspects can be woven into Java code to check whether object invariants are violated at runtime. Benefiting from AOP in separation of crosscutting concerns and weaving mechanisms, generated aspects can do the verification task whenever values of objects' attributes are changed. A Verification Aspect Generator (VAG) tool has been developed allowing the automatic generation of verifying aspects from the UML/OCL constraints. Read More: http://www.worldscientific.com/doi/abs/10.1142/S0218194011005281 %0 Journal Article %@ 2288-7202 %A Nguyen, Thu-Trang %A Nguyen, Thi-Hau %A Nguyen, Ha-Nam %A Nguyen, Duc-Nhan %A Choi, GyooSeok %D 2016 %F SisLab:1862 %J The International Journal of Advanced Culture Technology %N 1 %P 28-30 %T Detecting user status from smartphone sensor data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1862/ %V 4 %X Due to the high increment in usage and built-in advanced technology of smartphones, human activity recognition relying on smartphone sensor data has become a focused research area. In order to reduce noise of collected data, most of previous studies assume that smartphones are fixed at certain positions. This strategy is impractical for real life applications. To overcome this issue, we here investigate a framework that allows detecting the status of a traveller as idle or moving regardless the position and the direction of smartphones. The application of our work is to estimate the total energy consumption of a traveller during a trip. A number of experiments have been carried out to show the effectiveness of our framework when travellers are not only walking but also using primitive vehicles like motorbikes. %0 Conference Paper %A Nguyen, Tien Dung %A Nguyen, Huu Hiep Hai %A Le, Thanh Ha %B 2013: IEEE RIVF International Conference on Computing and Communication Technologies (RIVF2013) %C Hanoi, VIETNAM %D 2013 %F SisLab:790 %P 282-285 %T Trademark Image Retieval Based on Scale, Rotation, Translation Invariant Features %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/790/ %0 Conference Paper %A Nguyen, Tien Dung %A Nguyen, Huu Hiep Hai %A Le, Thanh Ha %B Computing and Communication Technologies, Research, Innovation, and Vision for the Future (RIVF), 2013 IEEE RIVF International Conference on %D 2013 %F SisLab:1931 %P 282-285 %T Trademark image retrieval based on scale, rotation, translation invariant features %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1931/ %0 Conference Paper %A Nguyen, Tri Thanh %A Ha, Quang Thuy %B 2014: the 2nd International Conference on Computer Science, Applied Mathematics and Applications (ICCSAMA) %C Budapest, Hungary %D 2014 %F SisLab:1303 %P 229-242 %T Novel Operations for FP-tree Data Structure and Their Applications %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1303/ %V 282 %0 Conference Paper %A Nguyen, Trinh Dong %A Truong, Anh Hoang %B 2017 4th NAFOSTED Conference on Information and Computer Science (NICS 2017) %C Hanoi, Vietnam %D 2017 %F SisLab:2714 %T A formal contract-based model for component-based real-time systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2714/ %0 Conference Paper %A Nguyen, Trung Duc %A Hoang, Xuan-Huan %B 2014: Hội thảo các chủ đề chọn lọc của CNTT %D 2014 %F SisLab:1187 %I Nhà xuất bản Khoa học và Kỹ thuật %P 182-187 %T Một chỉ số đánh giá số cụm mới cho thuật toán C-means mờ %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1187/ %0 Conference Paper %A Nguyen, Trung Hai %A Nguyen, Dai Tho %A University of Engineering and Technology, Vietnam National University, Hanoi, %B 2011 International Conference on Advanced Technologies for Communications %C Danang City, Vietnam %D 2011 %F SisLab:2376 %P 169-173 %T An Efficient and Message-Optimal Multicast Routing Protocol in Mobile Ad-Hoc Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2376/ %X This paper presents a novel protocol for multicasting in mobile ad-hoc networks named Spanning Tree for Multicasting (STM), which establishes and maintains the spanning tree for each multicast group using optimal control packets, without requiring any underlying unicast routing protocol or any pre-configured state of group. Based on the original idea of optimal maintenance of a spanning tree (OMST), we improve this algorithm considerably and propose our multicast protocol, as well as implement it as an extension of simulator NS-2. We compare STM with MAODV and PUMA using NS-2 through various scenarios, the results show that STM attains top packet delivery ratio, while keeping less overhead than others. As far as we know, our implementation is the very first implementation inspired by OMST in practice. %0 Conference Paper %A Nguyen, Tuan Phong %A Le, Anh Cuong %B The 2016 IEEE RIVF International Conference on Computing and Communication Technologies %C Hanoi, Vietnam %D 2016 %F SisLab:2401 %P 114-119 %T A hybrid approach to Vietnamese word segmentation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2401/ %X Word segmentation is the very first task for Vietnamese language processing. Word-segmented text is the input of almost other NLP tasks. This task faces some challenges due to specific characteristics of the language. As in many other Asian languages such as Japanese, Korean and Chinese, white spaces in Vietnamese are not always used as word separators and a word may contain one or more syllables. In this paper, we propose an efficient hybrid approach to detect word boundary for Vietnamese texts using logistic regression as a binary classifier combining with longest matching algorithm. First, longest matching algorithm is used to catch words that contain more than two syllables in input sentence. Next, the system utilizes the classifier to determine the boundary of 2-syllable words and proper names. Then, the predictions having low confidence conducted by the classifier are verified by a dictionary to get the final result. Our system can achieve an F-measure of 98.82% which is the most accurate result for Vietnamese word segmentation to the best of our knowledge. Moreover, the system also has a high speed. It can run word segmentation for nearly 34k tokens per second. %0 Conference Paper %A Nguyen, Tuan Phong %A Le, Anh Cuong %A Le, Nguyen Khoi %B SSC: the 2016 Student Scientific Contest %C Hanoi %D 2016 %F SisLab:1496 %T A hybrid approach for Vietnamese word segmentation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1496/ %0 Conference Paper %A Nguyen, Van Giap %A Vu, Viet Anh %A Le, Dinh Minh %A Phan, Xuan Hieu %B SSC: the 2016 Student Scientific Contest %C Hanoi %D 2016 %F SisLab:1497 %T Healthkee: a symptom checker and personalized health information service %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1497/ %0 Report %9 Technical Report %@ TR2017-FIT-004 %A Nguyen, Van Nam %D 2017 %F SisLab:2781 %I University of Engineering and Technologies %T A comparative performance evaluation of web servers %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2781/ %X This paper surveys the fundamental software designs of high performance web servers that are widely used today for communication networks. Given the fast growing demands of the Internet and wide-ranging intranets, it is essentially to fully and deeply understand which factors that mainly affect the web servers and how these systems can be designed to satisfy a maximum number of requests per second from multiple users. The paper presents empirical analysis of several popular webservers including Apache,NodeJSandNginXtopreciselyfigureoutthe trade-off of different software design to tackle the performance bottleneck and resource consuming problems. %0 Journal Article %@ 0866-8612 %A Nguyen, Van Nam %A Tran, Van Cuong %A Trần Văn Cường, %D 2016 %F SisLab:1994 %I VNU Journal of Science %J VNU Journal of Computer Science and Communication Engineering %N 2 %P 43-48 %T An Efficient Log Management System %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1994/ %V 32 %X Server monitoring is really necessary because this can help administrators to track users’ activities in order to improve user management ability, load balancing as well as to detect DDoS attacks. Usually, server monitoring is based on logging. However, logging system is always considered to be expensive in term of storage, data collection, data searching and analysis. In this paper, we introduce eLMS, an efficient and scalable log management system. In our system, the log files can be collected from multiple servers, stored in a scalable manner, appropriately indexed and fast analysed. eLMS acts both in online and off-line modes and provides a practical web-based monitoring interface. eLMS is based on a famous open-source core called ELK including three main modules: ElasticSearch, LogStash and Kibana. However, by using queue and streaming technique instead of uploading technique used in LogStash, eLMS is tested to be at least ten times faster than existing ELK %0 Conference Paper %A Nguyen, Van Nhat %A Nguyen, Ha Thanh %A Vo, Dinh Hieu %A Nguyen, Le Minh %B 2018 The 5th NAFOSTED Conference on Information and Computer Science (NICS) %C Ho Chi Minh city, Vietnam %D 2018 %F SisLab:3176 %T Relation Extraction in Vietnamese Text via Piecewise Convolution Neural Network with Word-Level Attention %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3176/ %X With the explosion of information technology, the Internet now contains enormous amounts of data, so the role of information extraction systems becomes very important. Relation Extraction is a sub-task of Information Extraction, which focuses on classifying the relationship between the entity pairs mentioned in the text. In recent years, despite the many new methods have been introduced, Relation Extraction still receives attention from researchers for languages in general and Vietnamese in particular. Relation Extraction can be addressed in a variety of ways, including supervised learning methods, unsupervised and semi-supervised methods. Recent studies in the English language have shown that Relation Extraction using deep learning method in the supervised or semi-supervised domains is achieving optimal and superior results over traditional non-deep learning methods. However, researches in Vietnamese are few and in the process of searching documents, the results of deep learning applying for Relation Extraction in Vietnamese are not found. Therefore, the research focuses on studying and research the method of using deep learning to solve Relation Extraction task in Vietnamese. In order to solve the Relation Extraction task, the research proposes and constructs a deep learning model named Piecewise Convolution Neural Network with Word-Level Attention. %0 Conference Paper %A Nguyen, Van Tham %A Nguyen, Ngoc Thanh %A Tran, Trong Hieu %B Intelligent Information and Database Systems - 11th Asian Conference, ACIIDS 2019, Proceedings, Part I %C Yogyakarta, Indonesia %D 2019 %F SisLab:3476 %P 3-15 %T Algorithms for Merging Probabilistic Knowledge Bases %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3476/ %0 Conference Paper %A Nguyen, Van Tham %A Nguyen, Ngoc Thanh %A Tran, Trong Hieu %B 2018 10th International Conference on Computational Collective Intelligence %C Bristol, UK %D 2018 %F SisLab:3188 %P 31-42 %T Framework for Merging Probabilistic Knowledge Bases %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3188/ %0 Journal Article %@ 10641246 %A Nguyen, Van Tham %A Nguyen, Ngoc Thanh %A Tran, Trong Hieu %D 2019 %F SisLab:3670 %I IOS Press %J Journal of Intelligent & Fuzzy Systems %P 1-14 %T A distance-based approach for merging probabilistic knowledge bases %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3670/ %X In the stages of development of probabilistic expert systems, knowledge merging is a major concern. To deal with knowledge merging problems, several approaches have been put forward. However, in the proposed models, each original probabilistic knowledge base (PKB) is represented by a set of probabilistic functions fulfilling such knowledge base. The drawbacks of the solutions are that the output of model is also a set of probabilistic functions satisfying the resulting PKB and there is no algorithm for implementing the merging process of PKBs in which each of them consists of probabilistic constraints. In this paper, distance-based approach is utilized to propose a new method of merging PKBs to ensure that both the input and output of methods are represented by sets of probabilistic constraints. To this aim, the relationship between the probability rules and the probabilistic constraints, and the several transformation methods for the representation of the original PKB are presented, a set of merging operators (MOs) is proposed, and several desirable logical properties are investigated and discussed. Several algorithms for merging PKBs are presented and the computational complexities of these algorithms are also analyzed and evaluated. %0 Journal Article %@ 0957-4174 %A Nguyen, Van Tham %A Nguyen, Ngoc Thanh %A Tran, Trong Hieu %F SisLab:4288 %I Elsevier %J Expert Systems with Applications %T A model for building probabilistic knowledge-based systems using divergence distances %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4288/ %X The knowledge-based systems (KBSs) in general and solving the knowledge merging problem in particular have seen a great surge of research activity in recent years. However, there still exist two main shortcomings that need to be addressed in the probabilistic framework. Firstly, the current methods only deal with the problems in which original probabilistic knowledge bases (PKBs) are required to be consistent and formed in the same structure. It is a very strong requirement and difficult to apply in practice. Secondly, only a few measures of distance between probability distributions have been studied to apply in existing models. To overcome these disadvantages, in this paper, we introduce a novel framework for merging PKBs. To this end, a theoretical model is introduced and several experiments are implemented. In theoretical model, some theorems are pointed out and proved to provide mathematical background to construct the merging model. Moreover, a deep survey on how to employ divergence distance functions (DDFs) between probability distributions to carry out the merging process are performed. In experimental aspect, a consistency recovery algorithm and some merging algorithms based on DDFs are proposed. Through the results of conducted experiments, issues about the time cost of merging process, the number of iterations, and CPU Time Elapsed parameter to solve the class of optimization problems in the merging process are analyzed, compared, and evaluated. %0 Journal Article %A Nguyen, Van Tham %A Nguyen, Ngoc Thanh %A Tran, Trong Hieu %A Nguyen, Do Kieu Loan %D 2018 %F SisLab:3192 %I Taylor & Francis %J Cybernetics and Systems %N 5-6 %P 317-338 %T Method for Restoring Consistency in Probabilistic Knowledge Bases %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3192/ %V 49 %0 Conference Paper %A Nguyen, Van Tham %A Tran, Trong Hieu %B The 2017 International Conference on Knowledge and Systems Engineering (KSE) %C Hue, Vietnam %D 2017 %F SisLab:2889 %I IEEE %T Inconsistency measures for probabilistic knowledge bases %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2889/ %X One of the major concerns in knowledge integration is inconsistencies in knowledge bases. An inconsistency measure is a tool that helps analyzing inconsistency knowledge bases and resolving inconsistencies. In recent years, a wide range of measures with desirable properties have been proposed, however, these measures often correspond to logical, or probabilistic-logical framework. In this paper, we investigate several inconsistency measures and their properties for the knowledge bases represented by probabilistic framework. %0 Conference Paper %A Nguyen, Van Tham %A Tran, Trong Hieu %B 9th International Conference on Knowledge and Systems Engineering %C Hue, Vietnam %D 2017 %F SisLab:3199 %P 148-153 %T Inconsistency measures for probabilistic knowledge bases %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3199/ %0 Conference Paper %A Nguyen, Van Tham %A Tran, Trong Hieu %B 2018 10th Asian Conference on Intelligent Information and Database Systems (ACIIDS) %C Dong Hoi, Vietnam %D 2018 %F SisLab:3191 %P 3-14 %T Solving Inconsistencies in Probabilistic Knowledge Bases via Inconsistency Measures %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3191/ %0 Conference Paper %A Nguyen, Van Toi %A Pham, Bao Son %A Phan, Xuan Hieu %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1539 %T Phát hiện và theo dõi các sự kiện truyền thông xã hội trực tuyến cho tiếng Việt %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1539/ %X Ngày nay, mạng internet cung cấp cho người dùng một lượng lớn thông tin và tri thức. Đặc biệt, trong những năm gần đây, số lượng người dùng mạng xã hội (social network) ngày càng tăng. Bởi sự tiện lợi, dễ dàng của nó mà những thông tin, sự kiện của thế giới thực thường được đăng tải trên các mạng xã hội ngay khi nó xảy ra. Trong khi đó các trang tin tức khác trên mạng thường đăng tải các thông tin này chậm hơn. Thậm chí, nhiều thông tin được đăng tải trên các trang mạng xã hội nhưng không được đăng tải qua các trang tin tức khác. Tin tức, thông tin, sự kiện sẽ có giá trị cao khi nó được truyền tải đến người dùng nhanh và chính xác. Vậy làm thế nào để tự động phát hiện, tập hợp nhanh các sự kiện đó và trả lời được các câu hỏi “sự kiện gì, xảy ra ở đâu? thời gian nào? diễn biến của sự kiện đó như thế nào?...” cho người dùng? Trong nghiên cứu này, chúng tôi muốn xây dựng một hệ thống phát hiện và theo dõi các sự kiện thông qua việc sử dụng dữ liệu là các tin tức (news) trên các trang mạng, các bình luận (comment), các bài (post) trên mạng xã hội Facebook và ngôn ngữ của các dữ liệu này là tiếng Việt. %0 Conference Paper %A Nguyen, Van Tu %A Le, Anh Cuong %A Nguyen, Ha Nam %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1538 %T Phân loại câu hỏi sử dụng nhiều loại đặc trưng %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1538/ %X Phân loại câu hỏi là một thành phần quan trọng trong các hệ thống hỏi đáp tự động. Nhiệm vụ chính của phân loại câu hỏi là dự báo kiểu thực thể của câu trả lời của các câu hỏi viết bằng ngôn ngữ tự nhiên. Phân loại câu hỏi có thể được thực hiện bằng các tiếp cận khác nhau như: tiếp cận dựa trên luật, tiếp cận dựa trên học máy. Các đặc trưng khác nhau về từ vựng, cú pháp và ngữ nghĩa có thể được trích xuất tự động từ các câu hỏi để phục vụ việc phân loại. Trong nghiên cứu này chúng tôi kết hợp các đặc trưng về từ vựng, cú pháp, ngữ nghĩa trong phân loại câu hỏi. Chúng tôi đề xuất sử dụng mẫu câu hỏi (Question pattern) như là một đặc trưng mới để kết hợp với các đặc trưng khác trong phân loại câu hỏi. Chúng tôi cũng đề xuất sử dụng các tập đặc trưng khác nhau cho mỗi nhóm câu hỏi với các từ để hỏi khác nhau. Chúng tôi nhận thấy rằng khi sử dụng mẫu câu hỏi như là một đặc trưng và kết hợp với các đặc trưng từ vựng, cú pháp, ngữ nghĩa khác có thể cải thiện đáng kể độ chính xác của phân loại câu hỏi. Chúng tôi đã kiểm tra những đề xuất của mình bằng cách sử dụng bộ phân loại Support Vector Machine trên bộ dữ liệu TREC và đã đạt được độ chính xác phân loại câu hỏi cao hơn so với những nghiên cứu trước đó trên cùng nguyên tắc phân loại và tập dữ liệu. %0 Journal Article %A Nguyen, Van Vuong %A Mai, Trong Nhuan %A Nguyen, Ngoc Thach %A Nguyen, Hai Chau %A Bui, Quang Hung %A Dao, Ngoc Thanh %D 2014 %F SisLab:1428 %J Tạp chí Khoa học, Đại học Quốc gia Hà Nội %T Khung cấu trúc hệ thống cơ sở dữ liệu liên ngành phục vụ phát triển bền vững vùng Tây Bắc, Việt Nam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1428/ %0 Conference Paper %A Nguyen, Van Vy %A Dang, Duc Hanh %B National Conference on Information Technology %C Vietnam %D 2003 %F SisLab:1969 %P 500-507 %T A Design Pattern of the Layer Interacting with Relational Databases in Object-Oriented Systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1969/ %0 Conference Paper %A Nguyen, Viet Anh %B ICERI 2016 Conference %C Seville, Spain %D 2016 %F SisLab:2393 %T Examining student’s satisfaction with online learning activities in blended learning course: a case study %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2393/ %0 Journal Article %@ 0219-6492 %A Nguyen, Viet Anh %D 2017 %F SisLab:2598 %I World Scientific %J Journal of Information & Knowledge Management %N 4 %P 1750040 %T The Impact of Online Learning Activities on Student Learning Outcome in Blended Learning Course %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2598/ %V 16 %X The aim of the study was to determine the impact of online learning activities to the learning outcomes of students who participated in the blended learning course. Interactive activities are considered, in this study, include teacher - student interaction, student - student interaction, student - content interaction, and student - technology interaction. The undergraduate student participated in the blended learning course which using formative assessment to evaluate student learning outcomes by the combination of different learning activities through a learning management system. The quantitative results obtained when implementing learning analytics data from the system through using regression analysis showed that the students interact effectively with learning activities in the course have better results. Quantitative analytical results indicate that student – student interaction has a greater impact on student learning outcomes. These learning activities used for interactive activities as suggestions for teachers to design and implement learning activities for blended learning courses %0 Journal Article %@ 2056-4880 %A Nguyen, Viet Anh %D 2017 %F SisLab:2396 %I emeraldinsight %J International Journal of Information and Learning Technology %N 1 %P 10-20 %T Towards the implenmentation of an assessment-centered blended learning framework at the course level: a case study in a Vietnamese national university %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2396/ %V 34 %0 Conference Paper %A Nguyen, Viet Anh %B The 8th International Conference on Frontiers of Educational Technologies %C Yokohama, Japan %F SisLab:4749 %T A model to detect student’s learning styles in the blended learning course %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4749/ %X Recently, personalized learning is becoming more and more popular, especially in blended learning courses. The student learning style is one of several factors to personalize content and appropriate learning activities for each learner. This article proposes a model to detect learning styles in blended learning courses to classify learners for personalization. The proposed model focuses on two phases: the online and face-to-face learning phases of the learning process. Besides, we also present several parameters to map resources and activities in a blended-learning course to a learning style model. Based on the identified criteria, the experimental results with 205 students' data when classifying learning styles by the Support Vector Machine method give an accuracy of 76.7% - 83.2% in the 04 dimensions of the Felder and Silverman model. Experimental results when applying two approaches: literature-based and driven-based, show that learners' styles are similar up to 83%. Findings show that the student's learning style does not change much in the learning process. As for course design implications, we also propose suggestions for developing content and designing learning activities. %0 Journal Article %@ 1360-2357 %A Nguyen, Viet Anh %D 2016 %F SisLab:2395 %I Springer %J Education and Information Technologies %T A peer assessment approach to project based blended learning course in a Vietnamese higher education %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2395/ %X This article presents a model using peer assessment to evaluate students taking part in blended - learning courses (BL). In these courses, teaching activities are carried out in the form of traditional face-to-face (F2F) and learning activities are performed online via the learning management system Moodle. In the model, the topics of courses are built as a set of projects and case studies for the attending students divided into groups. The result of the implementation of projects is evaluated and ranked by all course participants and is one of the course evaluation criteria for lecturers. To assess learners more precisely, we propose a multi-phase assessment model in evaluating all groups and the group members. The result of each student in the group based on himself evaluation, evaluations of the team members, the tearcher and all students in the course. There are 107 students, who participated in the course entitled ``web application development'', are divided into 20 groups conducting the course in the field of information technology is deployed in the form of blended learning through peer assessment. The results of student's feedback suggested that the usage of various peer assessment created positive learning effectiveness and more interesting learning attitude for students. The survey was conducted with the students through the questionnaire, each question with scale 5-point Likert scale that ranged from 1 (very unsatisfied) to 5 (very statisfied) to investigate the factors: Collaboration, Assessment, Technology showed that students were satisfied with our approach. %0 Journal Article %@ 2354-1059 %A Nguyen, Viet Anh %A Hoang, Manh Tien %A Nguyen, Hoa Huy %D 2016 %F SisLab:1760 %I Trường Đại học Sư phạm Hà nội %J Journal of Science of HNUE %N 4 %P 172-176 %T Mô hình chia sẻ tài nguyên học liệu trực tuyến %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1760/ %V 61 %X Bài báo giới thiệu mô hình chia sẻ tài nguyên trực tuyến cho phép cộng đồng người sử dụng tạo lập và chia sẻ các tài liệu học tập và giảng dạy, từ đó hình thành một kho học liệu mở với học liệu tuân thủ chuẩn scorm, dễ dàng tích hợp vào các hệ thống Learning Management System. Bên cạnh đó, chúng tôi cũng trình bày một số chức năng chính của phần mềm ConnectData do chúng tôi xây dựng để thử nghiệm mô hình. Kết quả thăm dò khi triển khai phần mềm ConnectData để thử nghiệm mô hình đã nhận được ý kiến phản hồi tích cực từ phía người sử dụng. %0 Conference Paper %A Nguyen, Viet Anh %A Le, Thanh Ha %A Nguyen, Thi Thuy %B The 7th International Symposium on Information and Communication Technology (SoICT) %C Ho Chi Minh city, Vietnam %D 2016 %F SisLab:2038 %T Single Camera Based Fall Detection Using Motion and Human Shape Features %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2038/ %0 Journal Article %A Nguyen, Viet Anh %A Nguyen, Hoa-Huy %A Nguyen, Duc-Loc %A Le, Minh-Duc %D 2021 %F SisLab:4571 %J Education and Information Technologies %T A course recommendation model for students based on learning outcome %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4571/ %X How to choose the most appropriate courses to study throughout the learning process remains a question interested in by many students. Students often choose suitable courses according to their interests, needs, and advice from supporting staff, etc. This paper presents the results in developing a course recommendation system that will select appropriate courses for each student studying a major in the following semesters based on his/her current academic performance. We have applied several techniques based on data mining and learning analytics to predict students’ learning outcomes in the next semester and developed a model to select the appropriate courses based on such a recommendation system. Besides, our study has focused on comparing the effectiveness of predictive learning methods based on collaborative filtering. Experiments analyzed the learning results of 510 students who enrolled in the courses from 2015 to 2019 and showed that the Matrix Factorization method is the most effective. Also, the paper has proposed procedures and constraints applicable to different training curricula. %0 Journal Article %@ 2377-2891 %A Nguyen, Viet Anh %A Nguyen, Manh Duy %D 2018 %F SisLab:3029 %J International Journal of Learning and Teaching %N 3 %P 178-184 %T An online question and answering system for support teacher-student interaction in the blended learning course %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3029/ %V 4 %X Student-teacher interaction plays a major role in teaching and learning in blended learning courses. During each lesson, students often do not have time to ask the instructor about the relevant issues of the subject that they want to answer. In addition, students often hesitate to talk directly with teachers about issues related to the content of the subject. The Question Answering (Q&A) system, therefore, is a useful tool to assist students and teachers to engage in interactive activities, especially in online courses. This research focuses on reviewing the state of the art of two respects as methods and techniques for developing question and answer systems, including issues of question classification, retrieval of information and extraction of answers as core components of the Q&A system. In addition, the paper proposes a model of an online questionnaire system for blended interactive learning activities which are being tested at a university training institution. Initially, the test results show that students are interested in this form of online question and answer. %0 Conference Paper %A Nguyen, Viet Anh %A Nguyen, Quang Bach %A Nguyen, Vuong Thinh %B The 2nd International Conference on E-Society, E-Education and E-Technology %D 2018 %F SisLab:3163 %P 35-41 %T A Model to Forecast Learning Outcomes for Students in Blended Learning Courses Based On Learning Analytics %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3163/ %0 Conference Paper %A Nguyen, Viet Ha %A Bui, The Duy %A Vu, Quang Dung %A Nguyen, Thi Thu Hien %B 4th International Conference on New Trends in Information Science and Service Science %C Gyeongju, Korea %D 2010 %F SisLab:1769 %P 210-213 %T Towards scalable agent based web service systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1769/ %X The increasing development of the Internet requires highly scalable, available, and reliable web servers in order to provide rapid response and high throughput to requests from hundreds millions users occurring at any time. The disadvantage of old fashioned approach which uses single-alone web servers is the problem of bottle-neck and overload on the server side as well as many additional problems for the network such as network traffic and routing overload. The load balancing problem of web servers can be solved with replicated web servers. However, the problems on the network are still left out. In this paper, we propose a framework to reallocate jobs from web servers to proxy servers that are close to the clients. By delivering certain jobs from requested web servers to designated proxy servers with mobile agents and executing these jobs at the proxy servers, our approach can reduce the load for web servers as well as reduce the traffic for the overall network. %0 Journal Article %A Nguyen, Viet Tan %D 2013 %F SisLab:758 %J Tạp chí khoa học và công nghệ của Viện Khoa học và Công nghệ Việt Nam %T Lọc cộng tác bằng phương pháp đồng huấn luyện %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/758/ %0 Journal Article %A Nguyen, Viet Tan %A Hoang, Vu %A Dang, Vu Tung %A Tu, Minh Phuong %D 2014 %F SisLab:1182 %J Tạp chí Khoa học Đại học Quốc gia Hà Nội %T Phân loại dữ liệu có liên kết sử dụng phương pháp đồng huấn luyện %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1182/ %0 Conference Paper %A Nguyen, Xuan Duc %A Le, Hong Quang %A Le, Viet Quan Trong %A Nguyen, Duc Nam %A Tran, Le Minh Nhat %A Pham, Bao Son %A Phan, Xuan Hieu %B SSC: the 2016 Student Scientific Contest %C Hanoi %D 2016 %F SisLab:1514 %T An automated system for monitoring online social media in educational area %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1514/ %0 Conference Paper %A Nguyen, Xuan Duc %A Le, Hong Quang %A Le, Viet Quan Trong %A Nguyen, Duc Nam %A Tran, Le Minh Nhat %A Pham, Bao Son %A Phan, Xuan Hieu %B SSC: the 2016 Student Scientific Contest %C Hanoi %D 2016 %F SisLab:1503 %T An automated system for monitoring online social media in educational area %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1503/ %0 Conference Paper %A Nguyen, Xuan Duc %A Nguyen, Minh Duc %A Tran, Mai Vu %A Phan, Xuan Hieu %A Pham, Bao Son %B 2017 4th NAFOSTED Conference on Information and Computer Science (NICS 2017) %C Hanoi, Vietnam %D 2017 %F SisLab:2690 %P 269-274 %T VNU-SMM: A social media monitoring framework on vietnamese online news %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2690/ %X The exponential growth of social media has created informative content, which leads to the emersion of social listening tools, in order to serve the need of companies and organizations. These tools are developed mainly in English language and for those which deal with Vietnamese, they lack the analysis capabilities to capture insightful information. In our research, we propose a framework to automatically extract and analyze information from social media, including online news and social networks. Our framework consists of multiple modules: data crawler, analysis module (topic analysis, sentiment analysis, event detection), visualization module. The results for analysis modules are promising: F1-score for topic analysis is 53%, average accuracy for sentiment analysis (using Maximum Entropy with three classes) is 62% and F1-score for event detection is 88.36%. %0 Conference Paper %A Nguyen, Xuan Nam %A Nguyen, Dai Tho %A University of Engineering and Technology, Vietnam National University, Hanoi, %B Hội thảo lần thứ II Một số vấn đề chọn lọc về an toàn an ninh thông tin %C Ho Chi Minh City, Vietnam %D 2017 %F SisLab:2759 %T Intrusion Detection Using a More General Feature Extraction Method for Payload-based Anomaly One-Class Classifier %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2759/ %X In this paper, we proposed a method to extract more general features of data for payload-based anomaly IDS. However, because of the significant rise in the number of features, there are numerous redundancies, leading to the rise in the complexity and the decrease in the accuracy of the classification. To that end, we apply Chi square [9] feature selection method to pick up the best features in the feature set. We have done many experiments on real world dataset of HTTP-based attacks to evaluate the performance of our classifier using our feature extraction method. The results show that our classifier can quickly detect the attack packets with very high true positive rate while keeping the false positive rate at a very low level. Besides, the results also indicate that our classifier outperforms other classifiers such as McPAD [10], and PAY [12, 13]. %Z Bài báo này đã giành được giải thưởng "Best Paper Award" tại hội thảo SOIS 2017. %0 Journal Article %A Nguyen, Xuan Nam %A Nguyen, Dai Tho %A University of Engineering and Technology, Vietnam National University, Hanoi, %D 2017 %F SisLab:2756 %I Bộ Thông tin và Truyền thông %J Chuyên san Các công trình nghiên cứu, phát triển và ứng dụng Công nghệ thông tin và Truyền thông %T Một phương pháp trích rút đặc trưng tổng quát hóa trong phát hiện thâm nhập mạng dựa trên bất thường %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2756/ %X Bài báo này đề xuất một phương pháp trích rút đặc trưng trong phân lớp dữ liệu tổng quát hơn so với một hệ thống phát hiện thâm nhập mạng dựa trên bất thường có tên là McPAD đã được đề xuất trước đây. Vấn đề đặt ra là số đặc trưng được trích rút tăng lên có thể làm tăng độ phức tạp và giảm độ chính xác của phương pháp phân lớp tổng quát hóa. Vì vậy, phương pháp lựa chọn đặc trưng Chi-Square được sử dụng để lọc ra chỉ những đặc trưng tốt nhất. Chúng tôi đã làm nhiều thí nghiệm với bộ dữ liệu là các tấn công HTTP thật để đánh giá hiệu năng của phương pháp trích rút đặc trưng tổng quát hóa kết hợp lọc Chi-Square. Kết quả cho thấy bộ phân lớp của chúng tôi có thể nhanh chóng phát hiện các gói tin tấn công với tỷ lệ dương tính thật rất cao trong khi duy trì tỷ lệ dương tính giả ở mức rất thấp. Ngoài ra, kết quả cũng cho thấy hiệu năng bộ phân lớp của chúng tôi vượt trội hơn so với McPAD. %Z Phiên bản tiếng Anh của bài báo này "Intrusion Detection Using a More General Feature Extraction Method for Payload-based Anomaly One-Class Classifier" đã được trình bày tại hội thảo lần thứ II Một số vấn đề chọn lọc về an toàn thông tin (SOIS 2017), qua phản biện được giải thưởng "Best Paper Award", được Cục An toàn thông tin của Bộ Thông tin và Truyền thông lựa chọn để dịch sang tiếng Việt đưa vào chuyên san Khoa học Công nghệ của Bộ Thông tin và Truyền thông số cuối năm 2017. %0 Conference Paper %A Nguyen, Xuan Nam %A Nguyen, Dai Tho %A Vu, Hai Long %A University of Engineering and Technology, Vietnam National University, Hanoi, %B 2016 3rd National Foundation for Science and Technology Development (NAFOSTED) Conference on Information and Computer Science (NICS) %C Danang City, Vietnam %D 2016 %F SisLab:2361 %P 74-79 %T POCAD: a Novel Payload-based One-Class Classifier for Anomaly Detection %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2361/ %X In this paper, we propose a novel Payload-based One-class Classifier for Anomaly Detection called POCAD, which combines a generalized 2v-gram feature extractor and a one-class SVM classifier to effectively detect network intrusion attacks. We extensively evaluate POCAD with real-world datasets of HTTP-based attacks. Our experiment results show that POCAD can quickly detect malicious payload and achieves a high detection rate as well as a low false positive rate. The experiment results also show that POCAD outperforms state of the art payload-based detection schemes such as McPAD [8] and PAYL [5]. %0 Journal Article %@ 1976-913X %A Nguyen, Xuan Thanh %A Ngo, Thi Duyen %A Le, Thanh Ha %D 2018 %F SisLab:3213 %I KIPS %J Journal of Information Processing Systems %T A Spatial-temporal 3D Human Pose Reconstruction Framework %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3213/ %X 3D human pose reconstruction from single-view camera is a difficult and challenging topic. Many approaches have been proposed, but almost focus on frame-by-frame independently while inter-frames are highly correlated in a pose sequence. In contrast, we introduce a novel spatial-temporal 3D reconstruction framework that leverages both intra and inter frame relationships in consecutive 2D pose sequences. Orthogonal Matching Pursuit (OMP) algorithm, pre-trained Pose-angle Limits and Temporal Models have been implemented. We quantitatively compare our framework with recent works on CMU motion capture dataset and Vietnamese traditional dance sequences. Our method outperforms others with 10 percent lower of Euclidean reconstruction error and robustness against Gaussian noise. Additionally, it is also important to mention that our reconstructed 3D pose sequences are smoother and more natural than others %0 Journal Article %@ 1976-913X %A Nguyen, Xuan Thanh %A Nguyen, Ba Tung %A Do, Khac Phong %A Bui, Quang Hung %A Nguyen, Thi Nhat Thanh %A Vuong, Van Quynh %A Le, Thanh Ha %D 2015 %F SisLab:1478 %I KIPS %J Journal of Information Processing Systems %N 1 %P 134-147 %T Spatial Interpolation of Meteorologic Variables in Vietnam using the Kriging Method %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1478/ %V 11 %0 Conference Paper %A Nguyen, Xuan Truong %A Pham, Manh Linh %B The 13th International Conference on Application of Information Technology in Agriculture Asia-Pacific Region (APFITA 2022) %C Hanoi, Vietnam %D 2022 %F SisLab:4790 %K Precision Livestock Farming, Precision Feeding System, health, cloud simulation %P 109-115 %T Cloud-Based Simulation of Precision Feeding System for Pig Health Management %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4790/ %0 Journal Article %@ 2250-2459 %A Nguyen, Xuan Truong %A Pham, Manh Linh %D 2022 %F SisLab:4789 %J International Journal of Emerging Technology and Advanced Engineering %K Precision feeding system, Simulation, Swine, Perturbations, Internet of Things %N 12 %P 136-145 %T Detecting Multiple Perturbations on Swine using Data from Simulation of Precision Feeding Systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4789/ %V 12 %0 Report %9 Technical Report %A Nguyen Binh, Nguyen %A Nguyen Van, Vinh %A VNU University of Engineering and Technology, %A VNU University of Engineering and Technology, %D 2018 %F SisLab:3350 %T Statistical Machine Translation For Vietnamese Grammatical Error Correction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3350/ %X Nowadays, along with the development of Natural Language Processing, there are a lot of research which use Statistical Machine Translation for grammatical error correction. Despite the fact that, there are a few researches which can be applied to Vietnamese. As a result, our purpose is to implement grammatical error correction in Vietnamese. The problem can easily describe like this: you have a wrong sentence as input, after being processed by the model, you will have the right sentence as output. In this research, we focus on applying Statistical Machine Translation to Vietnamese. This is a part of Machine Learning approach in order to solve the grammatical error correction problem. At first, we will try to create a list of all kind of Vietnamese’s error. Then, we aim for correcting simple error, like spelling error, then we develop the system step by step to handle and correct complex error. To do that, the model need lots of data to train, so we collect as much Vietnamese sentences as possible, and turn them into wrong to make parallel data. The data will be divided into three parts, which are used for training, tuning, and testing, respectively. After all, the model achieved some results, where the sentences with spelling mistake is corrected better than others. The result is not too good, but it can be seen that we can apply Statistical Machine Translation for the Grammatical error correction problem. %0 Journal Article %@ 2415-6698 %A Nguyen-Duc, Anh %A Do, Manh-Viet %A Luong-Hong, Quan %A Nguyen-Khac, Kiem %A Truong-Anh, Hoang %D 2021 %F SisLab:4609 %J Advances in Science, Technology and Engineering Systems Journal (ASTESJ) %N 2 %P 921-932 %T On the Combination of Static Analysis for Software Security Assessment – A Case Study of an Open-Source e-Government Project %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4609/ %V 6 %X Static Application Security Testing (SAST) is a popular quality assurance technique in software engineering. However, integrating SAST tools into industry-level product development and security assessment poses various technical and managerial challenges. In this work, we reported a longitudinal case study of adopting SAST as a part of a human-driven security assessment for an open-source e-government project. We described how SASTs are selected, evaluated, and combined into a novel approach for software security assessment. The approach was preliminarily evaluated using semi-structured interviews. Our result shows that (1) while some SAST tools outperform others, it is possible to achieve better performance by combining more than one SAST tools and (2) SAST tools should be used towards a practical performance and in the combination with triangulated approaches for human-driven vulnerability assessment in real-world projects. %0 Journal Article %A Nguyễn, Duy Phương %A Nguyen, Viet Tan %A Tu, Minh Phuong %D 2013 %F SisLab:772 %J Tạp chí Khoa học và Công nghệ, Viện Khoa học và Công nghệ Việt Nam %T Collaborative Filtering by CO-TRAINING Method %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/772/ %V 1A %0 Journal Article %A Nguyễn, Văn Hùng %A Nguyễn, Ngọc Linh %D 2018 %F SisLab:3243 %J Tạp chí KH&CN, Trường Đại học Công nghiệp %T NGHIÊN CỨU TÍCH HỢP MẠNG CẢM BIẾN KHÔNG DÂY DỰA TRÊN CÔNG NGHỆ ZIGBEE INTEGRATED SENSORS NETWORK RESEARCH BASE ON ZIGBEE TECHNOLOGY %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3243/ %V 49 %X Hiện nay, trên thế giới có rất nhiều chuẩn không dây (wireless) để truyền dữ liệu với tốc độ cao giữa các thiết bị với nhau như BlueTooth hay Wifi. Nhưng đối với những mạng quản lý các cảm biến trong các ứng dụng điều khiển - tự động hóa của các thiết bị trong nhà hay bệnh viện thì Wifi hay BlueTooth lại không thể đáp ứng được. Chúng có nhiều khuyết điểm như sử dụng băng thông rộng làm tiêu hao nhiều điện năng không cần thiết, sử dụng các nguồn điện trực tiếp, ít sử dụng pin, phạm vi kết nối nhỏ hẹp, độ trễ cao, cơ chế bảo mật đơn giản (BlueTooth), yêu cầu về các thiết bị phần cứng cao, chi phí lớn. Nhờ chức năng điều khiển từ xa không dây, truyền dữ liệu ổn định, tiêu thụ năng lượng cực thấp, công nghệ mở đã giúp công nghệ ZigBee trở nên hấp dẫn sử dụng cho các ứng dụng mạng không dây Trong nghiên cứu này, một mạng cảm biến không dây với công nghệ Zigbee ứng dụng trong nông nghiệp thông minh được tích hợp thử nghiệm với các đơn vị nhỏ nhất là các nút cảm biến có chức năng thu thập các thông số về môi trường, gửi đến trạm kiểm soát trung tâm và sau đó dữ liệu sẽ được gửi lên Database để quản lý và lưu trữ. Người dùng có thể theo dõi những thông số đó qua trang web hoặc điện thoại di động Android, từ đó đưa ra những quyết định phù hợp với cây trồng. Nhóm nghiên cứu đã thiết kế giao diện sao cho trực quan, dễ dàng sử dụng nhất để bất cứ ai cũng có thể tiếp cận được. Đây sẽ là một sản phẩm có thể được nhân rộng trên một vùng rộng lớn hoặc phạm vi tỉnh, thậm chí quốc gia, liên kết các vùng nông nghiệp trồng trọt lại với nhau, đưa nền nông nghiệp thông minh trở nên phổ biến và phát triển mạnh ở Việt Nam. %0 Conference Paper %A Nho, Minh Tu %A Nguyen, Hoa-Huy %A Ton, Quang Cuong %A Nguyen, Viet Anh %B First International Conference on Innovative Computing and Cutting-edge Technologies (ICICCT 2019) %C Istanbul, Turkey %D 2019 %F SisLab:3657 %T A Model to Forecast the Student's Grade and Course Recommendation: A Case Vietnamese Students %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3657/ %X This paper presents a model for forecasting the learning outcomes and suggesting the recommend courses for undergraduate students. In this research, we propose methods based on machine learning techniques and recommender systems to answer three research questions of forecasting student’s outcomes problem: 1) How to forecast the course’s grade for the next subject. 2) How to predict the final grade point average basing on the subjects students have studied. 3) How to suggest a list of subjects that students should learn in the next semester. The model has been tested with the grade data in 22-course subjects with the participation of 580 students. The results, in the best case, for predicting the missing subject grades with the tested data set were 0.656, a good result for scores in the range [0, 10]. Assessing the user’s satisfaction of the model through the survey, the results show that 68.2% of students think that the system is useful for them. %0 Journal Article %@ 1016-2364 %A Nong, Thi Hoa %A Bui, The Duy %D 2015 %F SisLab:1489 %I Institute of Information Science, Academia Sinica %J Journal of Information Science and Engineering %N 3 %P 713-726 %T An Improved Learning Rule for Fuzzy ART %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1489/ %V 30 %0 Conference Paper %A Nong, Thi-Hoa %A Bui, The Duy %B 2013: IEEE RIVF International Conference on Computing and Communication Technologies (RIVF2013) %C Hanoi, VIETNAM %D 2013 %F SisLab:783 %P 53-57 %T A Max-Min Learning Rule for Fuzzy ART %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/783/ %X Fuzzy Adaptive Resonance Theory (Fuzzy ART) is an unsupervised neural network, which clusters data effectively based on learning from training data. In the learning process, Fuzzy ARTs update the weight vector of the wining category based on the current input pattern from training data. Fuzzy ARTs, however, only learn from patterns whose values are smaller than values of stored patterns. In this paper, we propose a max-min learning rule of Fuzzy ART that learns all patterns of training data and reduces effect of abnormal training patterns. Our learning rule changes the weight vector of the wining category based on the minimal difference between the current input pattern and the old weight vector of the wining category. We have also conducted experiments on seven benchmark datasets to prove the effectiveness of the proposed learning rule. Experiment results show that clustering results of Fuzzy ART with our learning rule (Max-min Fuzzy ART) is significantly higher than that of other models in complex datasets. %0 Conference Paper %A Ogawa, Mizuhito %A To, Van Khanh %B 2013: the 20th Asia-Pacific Software Engineering Conference (APSEC 2013) %C Bangkok, Thailand %D 2013 %F SisLab:773 %T SAT and SMT: Their algorithm Designs and Applications %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/773/ %0 Conference Paper %A Ogbodo, Mark %A Dang, Nam Khanh %A Tomohide, Fukuchi %A Abdallah, Abderazek Ben %B The 2nd ACM Chapter International Conference on Educational Technology, Language and Technical Communication (ETLTC2020) %D 2020 %F SisLab:3955 %T Architecture and Design of a Spiking Neuron Processor Core Towards the Design of a Large-scale Event-Driven 3D-NoC-based Neuromorphic Processor %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3955/ %X Neuromorphic computing tries to model in hardware the biological brain which is adept at operating in a rapid, real-time, parallel, low power, adaptive and fault-tolerant manner within a volume of 2 liters. Leveraging the event driven nature of Spiking Neural Network (SNN), neuromorphic systems have been able to demonstrate low power consumption by power gating sections of the network not driven by an event at any point in time. However, further exploration in this field towards the building of edge application friendly agents and efficient scalable neuromorphic systems with large number of synapses necessitates the building of small-sized low power spiking neuron processor core with efficient neuro-coding scheme and fault tolerance. This paper presents a spiking neuron processor core suitable for an event-driven Three-Dimensional Network on Chip (3D-NoC) SNN based neuromorphic systems. The spiking neuron Processor core houses an array of leaky integrate and fire (LIF) neurons, and utilizes a crossbar memory in modelling the synapses, all within a chip area of 0.12mm2 and was able to achieves an accuracy of 95.15% on MNIST dataset inference. %0 Conference Paper %A Ouyang, Hua %A He, Niao %A Tran, Quoc Long %A Gray, Alexander %B International Conference on Machine Learning (ICML 2013) %D 2013 %F SisLab:3272 %P 80-88 %T Stochastic alternating direction method of multipliers %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3272/ %0 Journal Article %@ 1532-0626 %A Parlavantzas, Nikos %A Pham, Manh Linh %A Morin, Christine %A Arnoux, Sandie %A Beaunée, Gaël %A Qi, Luyuan %A Gontier, Philippe %A Ezanno, Pauline %D 2020 %F SisLab:3934 %I John Wiley & Sons Inc %J Concurrency and Computation: Practice and Experience %K Cloud computing; Epidemic simulation; High Performance Computing; Simulation models %N 5 %P 1-16 %T A Service-based Framework for Building and Executing Epidemic Simulation Applications in the Cloud %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3934/ %V 32 %X The cloud has emerged as an attractive platform for resource-intensive scientific applications, such as epidemic simulators. However, building and executing such applications in the cloud presents multiple challenges, including exploiting elasticity, handling failures, and simplifying multi-cloud deployment. To address these challenges, this paper proposes a novel service-based framework called DiFFuSE that enables simulation applications with a bag-of-tasks structure to fully exploit cloud platforms. This paper describes how the framework is applied to restructure two legacy applications, simulating the spread of bovine viral diarrhea virus and Mycobacterium avium subspecies paratuberculosis, into elastic cloud-native applications. Experimental results show that the framework enhances application performance and allows exploring different cost-performance trade-offs while supporting automatic failure handling and elastic resource acquisition from multiple clouds. %0 Conference Paper %A Pascal, Fontain %A Mizuhito, Ogawa %A Thomas, Sturm %A To, Van Khanh %A Vu, Xuan Tung %B Satisfiability Checking and Symbolic Computation %C United Kingdom %D 2018 %F SisLab:2975 %T EA: Wrapping Computer Algebra is Surprisingly Successful for Non-Linear SMT %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2975/ %X We report on a prototypical tool for Satisfiability Modulo Theory solving for quantifier-free formulas in Non-linear Real Arithmetic or, more precisely,real closed fields, which uses a computer algebra system as the main component. This is complemented with two heuristic techniques, also stemming from computer algebra, viz. interval constraint propagation and subtropical satisfiability. Our key idea is to make optimal use of existing knowledge and work in the symbolic computation community, reusing available methods and implementations to the most possible extent. Experimental results show that our approach is surprisingly efficient in practice. %0 Journal Article %@ 1663-4365 %A Peon, Antonio %A Dang, Cao Cuong %A Ballester, Pedro %D 2016 %F SisLab:2577 %J Frontiers in Chemistry %P 15 %T How Reliable Are Ligand-Centric Methods for Target Fishing? %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2577/ %V 4 %X Computational methods for Target Fishing (TF), also known as Target Prediction or Polypharmacology Prediction, can be used to discover new targets for small-molecule drugs. This may result in repositioning the drug in a new indication or improving our current understanding of its efficacy and side effects. While there is a substantial body of research on TF methods, there is still a need to improve their validation, which is often limited to a small part of the available targets and not easily interpretable by the user. Here we discuss how target-centric TF methods are inherently limited by the number of targets that can possibly predict (this number is by construction much larger in ligand-centric techniques). We also propose a new benchmark to validate TF methods, which is particularly suited to analyse how predictive performance varies with the query molecule. On average over approved drugs, we estimate that only five predicted targets will have to be tested to find two true targets with submicromolar potency (a strong variability in performance is however observed). In addition, we find that an approved drug has currently an average of eight known targets, which reinforces the notion that polypharmacology is a common and strong event. Furthermore, with the assistance of a control group of randomly-selected molecules, we show that the targets of approved drugs are generally harder to predict. %0 Conference Paper %A Pham, Canh V. %A Hoang, Xuan Huan %A Hoang M. Dinh, %A Hoan D. Nguyen, %A Huyen T. Dang, %B SoICT 2017 %C Nha Trang, Viet Nam %D 2017 %F SisLab:2764 %P 262-269 %T Limiting the Spread of Epidemics within Time Constraint on Online Social Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2764/ %X In this paper, we investigate the problem of limiting the spread of epidemics on online social networks (OSNs) with the aim to seek a set nodes of size at most k to remove from the networks such that the number of saved nodes is maximal for cases where we already know the set of infected nodes on the networks. The problem is proved to be NP-hard and it is NP-hard to approximate the problem with ratio nexp(1−ϵ), for 0 < ϵ < 1. Besides, we also suggest two algorithms to solve the problem. Experimental results show that our proposed outperform baseline algorithms. %0 Conference Paper %A Pham, Canh V. %A Phu, Quat V. %A Hoang, Xuan Huan %B 10th Asian Conference on Intelligent Information and Database Systems (AIIDS) %C Dong Hoi City, Vietnam %D 2018 %F SisLab:2790 %T Targeted Misinformation Blocking on Online Social Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2790/ %X In this paper, we investigate a problem of finding smallest set of nodes to remove from a social network so that influence reduction of misinformation sources at least given threshold γ, called Targeted Misinformation Blocking (TMB) problem. We prove that TBM is #P-hard under LT model. For any parameter � ∈ (0,γ), we designed Greedy algorithm which return the solution A with the expected influence reduction greater than γ−�, and the size of A is within factor 1+ln(γ/�) of the optimal size. To speed-up Greedy algorithm, we designed an efficient heuristic algorithm, called STBM algorithm. Experiments were conducted on real world networks which showed the effectiveness of proposed algorithms in term of both effectiveness and efficiency. %0 Conference Paper %A Pham, Chi Cong %A Trinh, Hai Thai %A Nguyen, Ngoc Hoa %B REV National Conference on Electronics, Communications and Information Technology %C Hanoi, Vietnam %D 2016 %F SisLab:2505 %T Phương thức quản lý truyền thông trong các hệ thống điều khiển và giám sát tập trung %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2505/ %X Trong các hệ thống điều khiển giám sát và thu thập dữ liệu (SCADA), bài toán quản lý truyền thông phải đối mặt với những thách thức về sự đa dạng trong sự truyền thông dữ liệu giữa các thiết bị công nghiệp. Chẳng hạn, từ việc hỗ trợ kết nối tới các RTU (Remote Terminal Unit) theo chuẩn truyền thông công nghiệp Modbus TCP/IP, OPC, … đến việc hỗ trợ các chuẩn mới như OPC-UA, Industrial IoT, … Ngoài ra, với sự đa dạng và phức tạp trong các nhà máy sản xuất công nghiệp, cũng cần phải có sự linh hoạt trong việc hỗ trợ các mô hình kết nối khác nhau, chẳng hạn như MultiClient, MultiServer, Redundancy. Từ thực trạng đó, nhóm nghiên cứu đã tiến hành khảo sát, đánh giá và đề xuất mô hình kiến trúc phân tầng và sử dụng chuẩn OPC-UA cho việc trao đổi dữ liệu trong các hệ thống thu thập và giám sát dữ liệu để phù hợp với yêu cầu trên. Mô hình này đã cho phép nâng cao khả năng quản lý truyền thông, cho phép mở rộng kết nối chỉ liên quan tới mở rộng giao diện tầng dưới mà không làm thay đổi cấu trúc các tầng phía trên. Thực nghiệm chúng tôi tiến hành trong khuôn khổ đề tài KC.03.26/11-15 đã cho phép minh chứng và khẳng định tính linh hoạt và hiệu quả của phương thức quản lý truyền thông đó của chúng tôi. %0 Journal Article %@ 0866-8612 %A Pham, Chinh Huu %A Le, Quoc Khanh %A Le, Thanh Ha %D 2014 %F SisLab:1915 %I VNU Journal of Science %J VNU Journal of Computer Science and Communication Engineering %N 3 %T Human action recognition using dynamic time warping and voting algorithm %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1915/ %V 30 %0 Conference Paper %A Pham, Dinh Chinh %A Miaomiao, Zhang %A Dang, Van Hung %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1541 %T On Model-Checking Probabilistic Timed Automata against Probabilistic Duration Properties %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1541/ %X In this paper, we consider a subclass of Probabilistic Duration Cal- culus formula called Simple Probabilistic Duration Calculus (SPDC) as a language for specifying dependability requirements for real-time systems, and address the two problems: to decide if a probabilistic timed automaton satisfies a SPDC formula, and to decide if there is a strategy to choose an execution of a given automaton that satisfies a SPDC formula. We prove that the both problems are decidable for a class of SPDC called probabilistic linear duration invariants, and provide a model checking algorithm for solving these problems. %0 Conference Paper %A Pham, Dinh Phong %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1542 %T Thiết kế ngữ nghĩa hình thang cho hệ phân lớp dựa trên luật mờ %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1542/ %X Hệ phân lớp dựa trên luật mờ được nghiên cứu rộng rãi do người dùng cuối có thể sử dụng những tri thức dạng luật được trích rút từ dữ liệu có tính dễ hiểu, dễ sử dụng đối với con người như là những tri thức của họ. Giải bài toán phân lớp dựa trên luật mờ là xây dựng bài toán trích rút tự động hệ dựa trên luật mờ từ dữ liệu sao cho hệ luật thu được phải dễ hiểu, tức là số luật và số điều kiện tạo nên luật phải đủ nhỏ, các giá trị ngôn ngữ của các từ ngôn ngữ xuất hiện trong hệ luật phải phản ánh được đúng ngữ nghĩa của bản thân chúng nhưng vẫn đáp ứng được yêu cầu đặt ra của bài toán. Đại số gia tử (ĐSGT) là một cách tiếp cận mới cho việc xử lý miền giá trị của biến ngôn ngữ, khai thác tính chất sánh được của các giá trị ngôn ngữ, là cơ sở xác định ngữ nghĩa định lượng từ ngữ nghĩa định tính, đã cho phép tạo ra các ràng buộc về ngữ nghĩa và đã được ứng dụng hiệu quả trong quá trình tìm kiếm, thiết kế tập giá trị ngôn ngữ cùng với ngữ nghĩa dựa trên tập mờ của chúng cho bài toán xây dựng tự động cơ sở luật cho hệ phân lớp dựa trên luật mờ. Tuy nhiên, với tiếp cận ĐSGT hiện tại, ngữ nghĩa lõi và ngữ nghĩa dựa trên tập mờ hình thang của các từ ngôn ngữ chưa được mô hình hóa. Nghiên cứu này đề xuất mở rộng lý thuyết ĐSGT biểu diễn được ngữ nghĩa của các từ ngôn ngữ phụ thuộc ngữ cảnh và đề xuất thay đổi phương pháp lượng hóa ĐSGT để mô hình hóa ngữ nghĩa định tính của các từ ngôn ngữ phù hợp với ngữ cảnh mới nhằm cung cấp một cơ chế hình thức cho việc sinh lõi ngữ nghĩa và ngữ nghĩa tính toán dựa trên tập mờ hình thang của khung nhận thức ngôn ngữ. Nghiên cứu này cũng chứng tỏ khả năng ứng dụng hiệu quả của ĐSGT mở rộng trong thiết kế tự động hệ phân lớp dựa trên luật mờ. %0 Conference Paper %A Pham, Duc Hong %A Le, Anh Cuong %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1543 %T A Least Square based Model for Rating Aspects and Identifying Important Aspects on Review Text Data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1543/ %X In recent years, opinion mining and sentiment analysis has been one of the attracting topics of knowledge mining and natural language processing. The problem of rating aspects from textual reviews is an important task in this field. In this paper we propose a new method for rating product aspects as well as for identifying important aspects in general. Our proposed model is based on the least square method and the QR decomposition technique. In our experiment, we use a dataset of 594810 reviews of 3775 hotels collected from the very famous website in tourism tripadvisor.com with five common aspects including cleanliness, location, service, room and value. Experimental result shows that our proposed method outperforms some well known studies for the same problem. %0 Journal Article %A Pham, Duc Hong %A Le, Anh Cuong %A Le, Hoan %D 2015 %F SisLab:1710 %I Học viện Kỹ thuật Quân sự %J Tạp chí Khoa học và Kỹ thuật %T Một mô hình hiệu quả đánh giá quan điểm người dùng đối với các thuộc tính sản phẩm sử dụng mạng Nơron %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1710/ %V 16 %0 Conference Paper %A Pham, Duc Hong %A Le, Anh Cuong %A Le, Thi Kim Chung %B NICS: 2nd National Foundation for Science and Technology Development Conference on Information and Computer Science %C Ho Chi Minh city, Vietnam %D 2015 %F SisLab:1461 %K data mining;least squares approximations;natural language processing;text analysis;cleanliness;hotel services;knowledge mining;least square based model;location;natural language processing;opinion mining;rating product aspects;review text data;room;sentiment analysis;service;value;Algorithm design and analysis;Computer science;Dictionaries;Hidden Markov models;Mathematical model;Prediction algorithms;Training %P 265-270 %T A Least Square based Model for Rating Aspects and Identifying Important Aspects on Review Text Data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1461/ %X Opinion mining and sentiment analysis has been one of the attracting topics of knowledge mining and natural language processing in recent years. The problem of rating aspects from textual reviews is an important task in this field. In this paper we propose a new method for rating product aspects as well as for identifying important aspects in general. Our proposed model is based on the least square method. The experiments are carried out on the data collected from hotel services with the aspects including the cleanliness, location, service, room, and value. We have obtained more accurate results than some well-known previous studies. %0 Conference Paper %A Pham, Huyen Trang %A Vu, Tien Thanh %A Tran, Mai Vu %A Ha, Quang Thuy %B Asia-Pacific Conference on Services Computing %D 2011 %F SisLab:2161 %P 503-508 %T A Solution for Grouping Vietnamese Synonym Feature Words in Product Reviews %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2161/ %0 Conference Paper %A Pham, Lien Vi %A Pham, Son Bao %A VNU-UET, %B 2012 Fourth International Conference on Knowledge and Systems Engineering (KSE) %C Danang, Vietnam %D 2012 %F SisLab:101 %K Data mining;Electronic mail;HTML;Information retrieval;Logic gates;Transducers;Internet;Web sites;advertising;information retrieval;real estate data processing;F-measure;Vietnamese online real estate advertisements;Vietnamese real estate Websites;information extraction;rule-based approach;information extraction;information extraction system;online real-estate advertising;real-estate;real-estate advertising; %P 181 -186 %T Information Extraction for Vietnamese Real Estate Advertisements %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/101/ %X Advertising has appeared in almost all areas of life. The large number of advertisements, especially in real estate domain, has raised a need for an effective way to search and find useful information for users. In this paper, we propose a rule-based approach to build an Information Extraction system for Vietnamese online real estate advertisements. Experimental results shows that our approach is very promising with an overall F-measure of 91 on a collection of data collected from popular Vietnamese real estate websites. %0 Journal Article %@ 1859-3534 %A Pham, Manh Linh %A Hoang, Tien Quang %A Nguyen, Xuan Truong %D 2021 %F SisLab:4443 %I Vietnam Ministry of Information and Communications %J Journal of Research and Development on Information and Communication Technology %K Elasticity, MQTT broker, Cloud computing, Internetof Things %N 2 %P 61-73 %T Elasticity for MQTT Brokers in IoT Applications %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4443/ %V 2020 %X Many domains of human life are more and more impacted by applications of the Internet of Things (i.e., IoT). The embedded devices produce masses of data day after day requiring a strong network infrastructure. The inclusion of messaging protocols like MQTT is important to ensure as few errors as possible in sending millions of IoT messages. This protocol is a great component of the IoT universe due to its lightweight design and low power consumption. Distributed MQTT systems are typically needed in actual application environments because centralized MQTT methods cannot accommodate a massive volume of data. Although being scalable decentralized MQTT systems, they are not suited to traffic workload variability. IoT service providers may incur expense because the computing resources are overestimated. This points to the need for a new approach to adapt workload fluctuation. Through proposing a modular MQTT framework, this article provides such an elasticity approach. In order to guarantee elasticity of MQTT server cluster while maintaining intact IoT implementation, the MQTT framework used off-the-shelf components. The elasticity feature of our framework is verified by various experiments. %0 Journal Article %@ 2588-1086 %A Pham, Manh Linh %A Hoang, Xuan Tung %D 2021 %F SisLab:4434 %J VNU Journal of Science: Computer Science and Communication Engineering %K MQTT broker, Elasticity, Internet of Things, Cloud computing %N 1 %P 1-14 %T An Elasticity Framework for Distributed Message Queuing Telemetry Transport Brokers %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4434/ %V 37 %X Internet of Things (IoT) applications are increasingly making impact in all areas of human life. Day by day, its chatty embedded devices have been generating tons of data requiring effective network infrastructure. To deliver millions of IoT messages back and forth with as few faults as possible, participation of communication protocols like Message Queuing Telemetry Transport (i.e., MQTT) is a must. Lightweight blueprint and battery friendly design are just two of many advantages of this protocol making it become a dominant in IoT world. In real application scenarios, distributed MQTT solutions are usually required since centralized MQTT approach is incapable of dealing with huge amount of data. Although distributed MQTT solutions are scalable, they do not adapt to fluctuations of traffic workload. This might cost IoT service providers because of redundant computation resources. This leads to the need of a novel approach that can adapt its volume changes in workload. This article proposes such an elastic solution by proposing a flexible MQTT framework. Our MQTT framework uses off-the-shelf components to obtain server’s elasticity while keeping IoT applications intact. Experiments are conducted to validate elasticity function provided by an implementation of our framework. %0 Journal Article %@ 1573-7543 %A Pham, Manh Linh %A Le, Nguyen Tuan Thanh %A Nguyen, Xuan Truong %D 2022 %F SisLab:4743 %I Springer Nature %J Cluster Computing %K Elasticity MQTT broker Internet of Things Cloud computing Smart homes %T Multi-level just-enough elasticity for MQTT brokers of Internet of Things applications %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4743/ %X Applications for the Internet of Things (IoT) are rapidly having an impact on all areas of daily life. Every day, its embedded devices generate loads of data that requires efficient network infrastructure. The integration of lightweight communication protocols such as Message Queuing Telemetry Transport (MQTT) is to send millions of IoT messages back and forth with as few errors as possible. In practice, IoT big data analytic systems are often deployed with highly regarded MQTT solutions to handle huge amounts of dynamic data and achieve scalability. However, these solutions do not adapt well to fluctuations in workload, so they are not elastic yet. This article introduces a novel framework that provides just-enough elasticity for MQTT brokers with multiple levels of virtualization and its implementation using EMQX MQTT broker, Kubernetes container-orchestration system and OpenStack cloud environment. Various experiments based on a real life IoT application are conducted to validate our proposed framework and its elastic functionality. %0 Conference Paper %A Pham, Manh Linh %A Nguyen, Ba Hieu %A Nguyen, Hoai Son %A Lê, Huy Hàm %B The 13th International Conference on KNOWLEDGE AND SYSTEMS ENGINEERING (KSE 2021) %C Bangkok, Thailand %D 2021 %F SisLab:4648 %K precision feeding system, simulation, agentbased modelling, swine, efficiency %T Simulation of precision feeding systems for swine %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4648/ %X Precision livestock farming has become an inevitable trend for agricultural industry in the world. In that field, precision feeding is widely acknowledged because of its potential to reduce feed costs, environmental footprint and to improve animal health and welfare. Precision feeding involves modern multidisciplinary technologies such as information technology, mechanics, electronics, automation, etc. Such a system consists of automatic troughs linked to a computer system to exploit data collected from the individual animals (e.g. body weight, feed intake and feeding behaviour), and/or from ambient sensors. Data is processed and analysed based on mathematical models to make predictions, warnings for farmers or to formulate diets that fit requirements of each individual animal at each production period. However, implementing such a system often requires high investment, which may go beyond the capabilities of smallholders and small/medium laboratories. Furthermore, the risk of implementing by design but not conforming to reality is very high. To avoid this problem, we introduce an agent-based modelling approach to simulate precision feeding systems for swine. Two simulation experiments were conducted to provide predictions about the growth of individual pigs and the usefulness of precision feeding systems over classic feeding models. %0 Journal Article %@ 1976-7277 %A Pham, Manh Linh %A Nguyen, Truong Thang %D 2019 %F SisLab:3442 %J KSII Transactions on Internet and Information Systems %K Application deployment, dynamic reconfiguration, distributed application, cloud computing, fog computing %N 3 %P 1141-1163 %T Flexible deployment of component-based distributed applications on the Cloud and beyond %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3442/ %V 13 %X In an effort to minimize operational expenses and supply users with more scalable services, distributed applications are actually going towards the Cloud. These applications, sent out over multiple environments and machines, are composed by inter-connecting independently developed services and components. The implementation of such programs on the Cloud is difficult and generally carried out either by hand or perhaps by composing personalized scripts. This is extremely error prone plus it has been found that misconfiguration may be the root of huge mistakes. We introduce AutoBot, a flexible platform for modeling, installing and (re)configuring complex distributed cloud-based applications which evolve dynamically in time. AutoBot includes three modules: A simple and new model describing the configuration properties and interdependencies of components; a dynamic protocol for the deployment and configuration ensuring appropriate resolution of these interdependencies; a runtime system that guarantee the proper configuration of the program on many virtual machines and, if necessary, the reconfiguration of the deployed system. This reduces the manual application deployment process that is monotonous and prone to errors. Some validation experiments were conducted on AutoBot in order to ensure that the proposed system works as expected. We also discuss the opportunity of reusing the platform in the transition of applications from Cloud to Fog computing. %0 Journal Article %@ 1530-8677 %A Pham, Manh Linh %A Nguyen, Truong Thang %A Hoang, Tien Quang %D 2021 %F SisLab:4605 %I Hindawi %J Wireless Communications and Mobile Computing %K fog computing, Internet of Things, elasticity, big data analytics, smart home %N 383364 %P 1-16 %T Towards an Elastic Fog Computing Framework for IoT Big Data Analytics Applications %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4605/ %V 2021 %X IoT applications have been being moved to the cloud during the last decade in order to reduce operating costs and provide more scalable services to users. However, IoT latency-sensitive big data streaming systems (e.g., smart home application) is not suitable with the cloud and needs another model to fit in. Fog computing, aiming at bringing computation, communication, and storage resources from “cloud to ground” closest to smart end-devices, seems to be a complementary appropriate proposal for such type of application. Although there are various research efforts and solutions for deploying and conducting elasticity of IoT big data analytics applications on the cloud, similar work on fog computing is not many. This article firstly introduces AutoFog, a fog-computing framework, which provides holistic deployment and an elasticity solution for fog-based IoT big data analytics applications including a novel mechanism for elasticity provision. Secondly, the article also points out requirements that a framework of IoT big data analytics application on fog environment should support. Finally, through a realistic smart home use case, extensive experiments were conducted to validate typical aspects of our proposed framework. %0 Journal Article %@ 2575-1719 %A Pham, Manh Linh %A Nguyen, Truong Thang %A Tran, Manh Dong %D 2019 %F SisLab:3588 %I Science Publishing Group %J International Journal of Information and Communication Sciences %K Benchmarking, MQTT, Cloud Computing, Internet of Things %N 4 %P 70-78 %T A Benchmarking Tool for Elastic MQTT Brokers in IoT Applications %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3588/ %V 4 %X Cloud computing is an evolution in IT consumption and delivery which makes available self-management on the Internet with a flexible, pay-as-you-go business model. Within the context of Internet of Things, the MQTT (Message Queuing Telemetry Transport) protocol that is implemented broadly by the applications of “Publish-Subscribe” paradigm has a vital role. However, MQTT brokers are saturated easily if they have to cope with huge and speedy data generated by IoT “chatty” devices. With capability of provisioning/deprovisioning granular virtual resources, Cloud computing empowered MQTT brokers by enabling its elasticity feature. Elasticity helps the brokers deal with a very large variety of data integrated into the IoT every single day. However, there was lack of sturdy benchmarking tools that judge all the aspects of MQTT brokers in order to advocate correct elastic decision-making. This article focuses on the work of benchmarking MQTT by introducing a new developed tool called MQTTBrokerBench. With this tool, users not only can benchmark MQTT brokers but also can specify saturation points where the IoT load makes the brokers be saturated. Those saturation points can be used to set thresholds for elastic decision-making. Furthermore, the article also demonstrates the results acquired by this tool through the experiments on Windows Azure Cloud Platform. %0 Journal Article %@ 2076-2615 %A Pham, Manh Linh %A Parlavantzas, Nikos %A Lê, Huy Hàm %A Bui, Quang Hung %D 2021 %F SisLab:4617 %I MDPI %J Animals %K veterinary epidemiology; African swine fever; high-performance simulation; modeling; transmission and spread %N 9 %P 2743 %T Towards a Framework for High-Performance Simulation of Livestock Disease Outbreak: A Case Study of Spread of African Swine Fever in Vietnam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4617/ %V 11 %X The spread of disease in livestock is an important research topic of veterinary epidemiology because it provides warnings or advice to organizations responsible for the protection of animal health in particular and public health in general. Disease transmission simulation programs are often deployed with different species, disease types, or epidemiological models, and each research team manages its own set of parameters relevant to their target diseases and concerns, resulting in limited cooperation and reuse of research results. Furthermore, these simulation and decision support tools often require a large amount of computational power, especially for models involving tens of thousands of herds with millions of individuals spread over a large geographical area such as a region or a country. It is a matter of fact that epidemic simulation programs are often heterogeneous, but they often share some common workflows including processing of input data and execution of simulation, as well as storage, analysis, and visualization of results. In this article, we propose a novel architectural framework for simultaneously deploying any epidemic simulation program both on premises and on the cloud to improve performance and scalability. We also conduct some experiments to evaluate the proposed architectural framework on some aspects when applying it to simulate the spread of African swine fever in Vietnam. %0 Conference Paper %A Pham, Manh Linh %A Tran, Manh Dong %A Nguyen, Truong Thang %B Hội thảo quốc gia lần thứ XXII: Một số vấn đề chọn lọc của Công nghệ thông tin và truyền thông %C Thái Bình, Việt Nam %D 2019 %F SisLab:3478 %K đo kiểm; điện toán đám mây; Internet vạn vật; MQTT %T Một công cụ đo kiểm giao thức MQTT cho những ứng dụng IoT phục vụ cho việc ra quyết định co dãn tài nguyên trên đám mây %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3478/ %X Trong sự phát triển của Internet vạn vật (IoT), giao thức MQTT (Message Queuing Telemetry Transport) có một vai trò rất quan trọng bởi lẽ nó được sử dụng rộng rãi bởi những ứng dụng IoT thường được triển khai theo mô hình "Xuất bản-Đăng ký". Với khả năng cung ứng và giải phóng những tài nguyên ảo và mịn (chức năng co dãn - elasticity), điện toán đám mây có khả năng làm gia tăng sức mạnh cho những máy chủ môi giới MQTT. Khả năng co dãn giúp các máy chủ môi giới MQTT có thể đương đầu với một khối lượng lớn dữ liệu đang từng ngày được tích hợp vào IoT. Tuy nhiên vẫn có sự thiếu vắng của các công cụ đo kiểm có thể đánh giá đầy đủ tất cả các khía cạnh của MQTT để hỗ trợ việc ra những quyết định co dãn tài nguyên hợp lý và chính xác. Bài báo này tập trung vào chủ đề đo kiểm MQTT thông qua việc giới thiệu MQTTBench một công cụ kiểm thử mới được chúng tôi phát triển cùng với những kết quả ban đầu có được từ việc áp dụng công cụ đo kiểm đó trên một ứng dụng điện toán đám mây giả lập. %0 Conference Paper %A Pham, Minh Duc %A Ly, Van Tuan %A Le, Dinh Thanh %B The 4th NAFOSTED Conference on Information and Computer Science (NICS) %C Hanoi, Vietnam %D 2017 %F SisLab:2640 %T Identification of Paths and Parameters in RESTful URLs for the Detection of Web Attacks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2640/ %X Attribute-specific models are quite effective in the detection of web attacks. Nevertheless, none of existing attribute-specific models can be applied directly to modern web applica-tions, which are dominant today and which use RESTful URLs. The barrier is that paths and parameters are mixed in URLs so that detection systems cannot identify which programs and attributes are which. To remove this barrier, this paper is the first one dealing with the identification of paths and parameters in RESTful URLs. The problem is raised, then a simple yet effective method to solve the problem is proposed. We validate our method via experiments. %0 Journal Article %@ 0866-8612 %A Pham, Nam Thanh %A Vu, Khuong Duy %A Dinh, Trieu Duong %A Le, Thanh Ha %D 2015 %F SisLab:1659 %I VNU Journal of Science %J VNU Journal of Computer Science and Communication Engineering %T Efficient Region-of-Interest Based Adaptive Bit Allocation for 3D-TV Video Transmission over Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1659/ %X Due to characteristics of human visual system (HVS), people usually focus more on a specific region named region-of-interest (ROI) of a video frame, rather than watch the whole frame. In addition, ROI-based video coding can also help to effectively reduce the number of encoding bitrates required for video transmission over networks, especially for the 3D-TV transmissions. Therefore, in this work, we propose a novel ROI-based bit allocation (BA) method which can adaptively extract and increase the visual quality of ROI while saving a huge number of encoding bitrates for video data. In the proposed method, we first detect and extract ROI based on the depth information obtained from 3D-TV video coding sequences. Then, based on the extracted ROI, a novel BA scheme is performed to solve the rate-distortion (R-D) optimization problem, in which the higher priority bitrates are adaptively assigned to ROI while the total encoding bitrates of video frames are kept satisfying all constraints required by the R-D optimization. Experimental results show that the proposed method provides much better higher peak signal-to-noise ratio (PSNR) as compared to other conventional BA methods. %0 Conference Paper %A Pham, Nghia Luan %A Nguyen, Van Vinh %B Paclic 2019 %D 2019 %F SisLab:3870 %T Adapting Neural Machine Translation for English-Vietnamese using Google Translate system for Back-translation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3870/ %X In this paper, we propose a method to better leveraging monolingual data by exploiting the advantages of GNMT system. Our method for adapting a general neural machine translation system to a specific domain, by exploiting Back-translation technique using targetside monolingual data. This solution requires no changes to the model architecture from a standard NMT system. Experiment results show that our method can improve translation quality, results significantly outperforming strong baseline systems, our method improves translation quality in legal domain up to 13.65 BLEU points over the baseline system for English-Vietnamese pair language. %0 Conference Paper %A Pham, Nghia Luan %A Nguyen, Van Vinh %A Nguyen, Huy Hoang %B Kỷ yếu Hội nghị Quốc gia lần thứ XII về Nghiên cứu cơ bản và ứng dụng Công Nghệ thông tin (FAIR) %C ĐH Khoa học - ĐH Huế %D 2019 %F SisLab:3876 %T THÍCH ỨNG MIỀN TRONG DỊCH MÁY NƠ RON CHO CẶP NGÔN NGỮ ANH - VIỆT %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3876/ %X Trong bài báo này, chúng tôi đề xuất một phương pháp thích ứng miền mới cho dịch máy nơ ron, áp dụng cho cặp ngôn ngữ Anh - Việt. Ý tưởng chính của bài báo là kết hợp dữ liệu đơn ngữ ngoài miền ở ngôn ngữ nguồn (tiếng Anh) với bản dịch của nó ở ngôn ngữ đích (tiếng Việt) để làm dữ liệu huấn luyện hệ dịch. Các thực nghiệm đã chứng minh rằng phương pháp chúng tôi đề xuất dễ thực hiện, khai thác được những ưu điểm của dữ liệu đơn ngữ như luôn có sẵn, chi phí xây dựng thấp và đặc biệt là chất lượng của hệ dịch được và tăng 2,21 điểm BLEU trong thử nghiệm của chúng tôi %0 Conference Paper %A Pham, Nghia Luan %A Nguyen, Van Vinh %A Nguyen, Quang Huy %B The 31st Pacific Asia Conference on Language, Information and Computation PACLIC 31 (2017) %D 2017 %F SisLab:2788 %T Translation model adaptation for Statistical Machine Translation with domain classifier %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2788/ %0 Conference Paper %A Pham, Ngoc Hung %A Nguyen, Viet Ha %A Aoki, Toshiaki %A Katayama, T. %B 2012 IEEE RIVF International Conference on Computing and Communication Technologies %D 2012 %F SisLab:34 %K assume-guarantee verification, complexity, component-based software verification, computational cost, consumption memory, depth-limited search strategy, minimized assumption generation, model checking, object-oriented programming, optimization, program verification, search tree, software evolution, software maintenance, state space explosion problem, subtree, system rechecking, tree searching %P 1 -6 %T An Improvement of Minimized Assumption Generation Method for Component-Based Software Verification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/34/ %0 Conference Paper %A Pham, Ngoc Hung %A Nguyen, Viet Ha %A Aoki, Toshiaki %A Katayama, T. %B 2012 IEEE RIVF International Conference on Computing and Communication Technologies, Research, Innovation, and Vision for the Future (RIVF) %C Ho Chi Minh city %D 2012 %F SisLab:111 %K Complexity theory;Computational efficiency;Context;Memory management;Search problems;Software;Software algorithms;object-oriented programming;program verification;software maintenance;tree searching;assume-guarantee verification;complexity;component-based software verification;computational cost;consumption memory;depth-limited search strategy;minimized assumption generation;model checking;optimization;search tree;software evolution;state space explosion problem;subtree;system rechecking; %P 1 -6 %T An Improvement of Minimized Assumption Generation Method for Component-Based Software Verification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/111/ %X The minimized assumption generation has been recognized as an improved method of the assume-guarantee verification for generating minimal assumptions. This method is not only fitted to component-based software but also has a potential to solve the state space explosion problem in model checking. However, the computational cost for generating the minimal assumption is very high so the method is difficult to be applied in practice. This paper presents an optimization as a continuous work of the minimized assumption generation method in order to reduce the complexity of the method. The key idea of this method is to find a smaller assumption in a sub-tree of the search tree containing the candidate assumptions using the depth-limited search strategy. With this approach, the improved method can generate smaller assumptions with a lower computational cost and consumption memory than the minimized method. The generated assumptions are also effective for rechecking the systems at much lower computational cost in the context of software evolution. We have implemented a tool supporting the improved method. Experimental results are also presented and discussed. %0 Journal Article %@ 0916-8532 %A Pham, Ngoc Hung %A Nguyen, Viet Ha %A Toshiaki, Aoki %A Takuya, Katayama %D 2010 %F SisLab:2039 %I IEICE %J IEICE Transactions on Information and Systems %N 8 %P 2172-2181 %T A Minimized Assumption Generation Method for Component-Based Software Verification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2039/ %V E93-D %0 Journal Article %@ 0916-8508 %A Pham, Ngoc Hung %A Nguyen, Viet Ha %A Toshiaki, Aoki %A Takuya, Katayama %D 2012 %F SisLab:2041 %I IEICE %J IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer %N 9 %P 1451-1460 %T On Optimization of Minimized Assumption Generation Method for Component-Based Software Verification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2041/ %V 95-A %0 Journal Article %@ 0916-8508 %A Pham, Ngoc Hung %A toshiaki, aoki %A takuya, katayama %D 2009 %F SisLab:2040 %I IEICE %J IEICE TRANSACTIONS on Fundamentals of Electronics, Communications and Computer %N 11 %P 2772-2780 %T Modular Conformance Testing and Assume-Guarantee Verification for Evolving Component-Based Software %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2040/ %V E92-A %0 Report %9 Technical Report %@ TR2018-FIT-11 %A Pham, Nguyen Hoang %A Luong, Ngoc Huyen %A Nguyen, Thi Hanh %A Dang, Duc Hanh %D 2018 %F SisLab:3198 %T Integrating a Testing Technique into the RTL Tool %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3198/ %0 Conference Paper %A Pham, Quoc Thang %A Nguyen, Duc Dung %A Nguyen, Thanh Thuy %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1544 %T The vector machine learning methods on sign language recognition problem %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1544/ %X Human gesture recognition is a rather new field and many challenges, especial when motion capture devices become more popular. Sign language recognition is a concrete example of gesture recognition. Various studies have shown that the vector machine methods with Gaussian kernels are among the most prominent models for an accurate gesture classification. In this study, we present the application of vector machine learning methods to sign language recognition problem. We demonstrate that the vector machines (VMs) could also achieve the state-of-the-art predictive performance. The experimental results on the Auslan data set show the feasibility and effectiveness of these methods. %0 Conference Paper %A Pham, Thanh Huyen %A Ho, Thuan %A Ha, Quang Thuy %A Nguyen, Tri Thanh %A Nguyen, Xuan Bach %B CHƯƠNG TRÌNH HỘI THẢO QUỐC GIA LẦN THỨ XXIII - MỘT SỐ VẤN ĐỀ CHỌN LỌC CỦA CÔNG NGHỆ THÔNG TIN VÀ TRUYỀN THÔNG %D 2020 %F SisLab:3712 %T Một mở rộng tập thô phủ và ứng dụng %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3712/ %X Lý thuyết tập thô và các mở rộng của nó như tính toán hạt dựa trên tập thô, tập thô phủ, tâp thô mờ,... ngày càng được phát triển và áp dụng hiệu quả trong nhiều lĩnh vực. Bài báo tập trung giới thiệu khái niệm bảng quyết định phủ, một ví dụ và phân tích về việc xây dựng bảng quyết định phủ. Bài báo cũng đưa ra một gợi ý về việcc ứng dụng bảng quyết định phủ trong hệ tư vấn. %0 Conference Paper %A Pham, Thanh Huyen %A Nguyen, Thi Cam Van %A Vuong, Thi Hong %A Ho, Thuan %A Ha, Quang Thuy %A Nguyen, Tri Thanh %A Trường Đại học Công nghệ, ĐHQGHN, %B 11th International Conference on Information Science and Applications 2020 (ICISA2020) %C Virtual Conference %D 2020 %F SisLab:4174 %T A Definition of Covering Based Decision Table and Its Sample Applications %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4174/ %X Covering based rough set, an extension of the traditional rough set theory, which uses the cover set of the universe set instead of the partition of the universe, has proven to be both theoretical and attractive in terms of applications. Corresponding to the decision table in traditional rough set theory, the concept of covering decision system has been defined. In this paper, we propose a decision table type based on covers, including the condition lattice of covers, and the decision lattice of covers. Two tasks on covering based decision table are also introduced. We also demonstrate the applications of the covering based decision table in collaborative filtering that corresponds to the classification in the traditional decision table, and in constraint based association rule mining to indicate this covering decision table concept has a potential application %0 Conference Paper %A Pham, Thanh Huyen %A Pham, Thi Ngan %A Ho, Thuan %A Vuong, Thi Hong %A Nguyen, Tri Thanh %A Ha, Quang Thuy %B ACIIDS 2021: Intelligent Information and Database Systems %F SisLab:4549 %T Solving reduction problems in cover lattice based decision tables %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4549/ %X Covering based rough set is an important extension of Paw-lak's traditional rough set. Reduction is a typical application of rough sets, including traditional, cover-ing based and other rough set extensions. Although this task has several pro-posals, it is still an open problem to decision systems (tables) in covering based rough set. This paper focuses on the reduction problem for the condition lattice, and fitting problem for the decision lattice in the decision table based on cover lat-tice. A corresponding algorithm is proposed for each problem. Two examples to illustrate a covering based decision table and two related problems show the ap-plications of these concepts and problems %0 Conference Paper %A Pham, Thanh Tung %A Dinh, Trieu Duong %A Hoang, Van Xiem %A Vu, Huu Tien %A Le, Thanh Ha %B The 10th International Conference on Knowledge and Systems Engineering (KSE2018) %D 2018 %F SisLab:3290 %T Coding distortion modelling method for local image perception %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3290/ %0 Conference Paper %A Pham, Thanh Tung %A Dinh, Trieu Duong %A Hoang, Van Xiem %A Vu Huu, Tien %A Le, Thanh Ha %B http://www.icce-asia2019.org/ %C Bangkok, Thailand %D 2019 %F SisLab:3457 %P 1-4 %T Distortion Model based on Perceptual of Local Image Content %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3457/ %X As humans are the ultimate receivers of the majority of visual signals being processed, the most accurate way of assessing image quality is to ask humans for their opinions of an image’s quality, known as the subjective visual quality assessment (VQA). The subjective image quality scores gathered from all subjects are processed to be the mean opinion score (MOS), which is regarded as the ground truth of image quality. Due to the fact that the human visual system (HVS) is differently sensitive to features of image patch, a novel coding distortion modelling method for local image perception is proposed in this paper. An experimental quality assessment to approach database for image patch has been developed. Mean opinion score is regarded as an essential parameter meanwhile the QP-MOS sigmoid curve is determined by local image content. %0 Conference Paper %A Pham, Thanh-Huyen %A Van-Tuan, Phan %A Thi-Ngan, Pham %A Thi-Hong, Vuong %A Tri-Thanh, Nguyen %A Quang-Thuy, Ha %B 14th Asian Conference on Intelligent Information and Database Systems %C Ho Chi Minh City, Vietnam %F SisLab:4751 %T A multi-label classification framework using the covering based decision table %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4751/ %X Multi-label classification (MLC) has recently drawn much attention thanks to its usefulness and omnipresence in real-world applications, in which objects may be characterized by more than one labels. One of the challenges in MLC is to deter-mine the relationship between the labels due to the fact that there is not any as-sumptions of the independence between labels, and there is not any information and knowledge about these relationships in a training dataset. Recently, many re-searches have focused on exploiting these label relationships to enhance the per-formance of the classification, however there have not many of them using the covering rough set. This paper propose a multi-label classification algorithm named CDTML, based on ML-KNN algorithm, using covering based decision table which exploits the relationship between labels to enhance the performance of the multi-label classifier. The experimental results on serveral dataset of Enron, Medical and a Vietnamese dataset of hotel reviews shown the effectiveness of the proposed algorithm. %0 Journal Article %@ 1873-4294 %A Pham, The Hai/V %A Nguyen, Hai Nam %A Doan, Viet Nga %A Dang, Thanh Hai %A Dieguez-Santana, Karel %A Marrero-Poncee, Yovani %A Castillo-Garit, Juan/A %A Casanola-Martin, Gerardo/M %A Le, Thi Thu Huong %D 2017 %F SisLab:2903 %I Bentham Science %J Current Topics in Medicinal Chemistry %P 1-20 %T Learning from Multiple Classifier Systems: Perspectives for Improving Decision Making of QSAR Models in Medicinal Chemistry %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2903/ %V 18 %X Quantitative Structure - Activity Relationship (QSAR) modeling has been widely used in medicinal chemistry and computational toxicology for many years. Today, as the amount of data on chemicals is increasing dramatically, QSAR methods have become pivotal for the purpose of handling the data, identifying a decision, and gathering useful information from data processing. The advances in this field have paved a way for numerous alternative approaches that require deep mathematics in order to enhance the learning capability of QSAR models. One of these directions is the use of Multiple Classifier Systems (MCSs) that potentially provide a means to exploit the advantages of manifold learning through decomposition frameworks, while improving generalization and predictive performance. In the present paper, we present MCS as a next generation of QSAR modeling techniques and discuss the chance to mining the vast number of models already published in the literature. We systematically revisited the theoretical frameworks of MCS as well as current advances in MCS application for QSAR practice. Furthermore, we illustrate our idea by describing ensemble approaches on modeling histone deacetylase (HDACs) inhibitors. We expect that our analysis would contribute to a better understanding about MCS application and its future perspectives for improving the decision making of QSAR models. %0 Conference Paper %A Pham, Thi Kim Dung %A Dubois, Catherine %A Levy, Nicole %B FMSPLE: the 6th Workshop on Formal Methods and Analysis in SPL Engineering %C London, United Kingdom %D 2015 %F SisLab:1648 %P 44-55 %T Towards correct-by-construction product variants of a software product line: GFML, a formal language for feature modules %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1648/ %V 182 %0 Conference Paper %A Pham, Thi Ngan %A Ha, Quang Thuy %A Nguyen, Minh Chau %A Nguyen, Tri Thanh %B The 6th International Conference on Computer Science, Applied Mathematics and Applications (ICCSAMA 2019) %D 2019 %F SisLab:3633 %T A probability-based close domain metric in lifelong learning for multi-label classification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3633/ %X Lifelong machine learning has recently become a hot topic attracting the re-searchers all over the world by its effectiveness in dealing with current problem by exploiting the past knowledge. The combination of topic modeling on pre-vious domain knowledge (such as topic modeling with Automatically generat-ed Must-links and Cannot-links, which exploits must-link and cannot-link of two terms), and lifelong topic modeling (which employs the modeling of previous tasks) is widely used to produce better topics. This paper proposes a close domain metric based on probability to choose reliable (prior) knowledge learnt from the past to generate more coherent topics on the current domain. This knowledge is, then, used to enrich features for multi-label classifier. Several experiments performed on review dataset of hotel show that the proposed approach leads to an improvement in performance over the baseline. %0 Conference Paper %A Pham, Thi Ngan %A Nguyen, Le Minh %A Ha, Quang Thuy %B IALP: International Conference on Asian Language Processing %C Hanoi %D 2012 %F SisLab:81 %P 85-89 %T Named Entity Recognition for Vietnamese Documents Using Semi-supervised Learning Method of CRFs with Generalized Expectation Criteria %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/81/ %0 Conference Paper %A Pham, Thi Ngan %A Nguyen, Van Quang %A Dinh, Duc Trong %A Nguyen, Tri Thanh %A Ha, Quang Thuy %B The 9th Asian Conference on Intelligent Information and Database Systems %C Kanazawa, Japan %D 2017 %F SisLab:2316 %I Springer %T MASS: a semi-supervised multi-label classification algorithm with specific features %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2316/ %X Multi-Label Classification (MLC), which, recently, has attracted several attentions, aims at building classification models for objects assigned with multiple class labels simultaneously. Existing approaches for MLC mainly focus on improving supervised learning which needs a relatively large amount of labeled training data. In this paper, we propose a semi-supervised algorithm to exploit unlabeled data for MLC for enhancing the performance. In the training process, our algorithm exploits the specific features per prominent class label chosen by a greedy approach as an extension of LIFT algorithm, and unlabeled data consumption mechanism from TESC. In classification, the 1-Nearest-Neighbor (1NN) is applied to select appropriate class labels for a new data instance. Our experimental results on a data set of hotel (for tourism) reviews indicate that a reasonable amount of unlabelled data helps to increase the F1 score. Interestingly, with a small amount of labelled data, our algorithm can reach comparative performance to a larger amount of labelled data. %0 Journal Article %A Pham, Thi Ngan %A Nguyen, Van Quang %A Tran, Van Hien %A Nguyen, Tri Thanh %A Ha, Quang Thuy %D 2017 %F SisLab:2504 %J Journal of Information and Telecommunication %T A semi-supervised multi-label classification framework with feature reduction and enrichment %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2504/ %X Multi-label classification has drawn much attention thanks to its usefulness and omnipresence in real-world applications in which objects may be char-acterized by more than one label as in the traditional approach. Getting mul-ti-label examples is costly and time-consuming therefore semi-supervised learning approach should be considered to take advantages of both labeled and unlabeled data. In this work, we propose a semi-supervised multi-label classification algorithm exploiting the specific features of the prominent class label(s) chosen by a greedy approach as an extension of the LIFT algo-rithm, and unlabeled data consumption mechanism from the TESC algo-rithm. We also make a semi-supervised multi-label classification application framework for Vietnamese texts with several feature enrichment steps in-cluding a) a stage of enriching features by adding hidden topic features; b) a stage of dimensional reduction for subtracting irrelevant features. Experi-mental results on a dataset of hotel reviews (for tourism) indicate that a rea-sonable amount of unlabeled data helps to increase the F1 score. Interesting-ly, with a small amount of labeled data, our algorithm can reach a compara-tive performance to the case of using a larger amount of labeled data. %0 Conference Paper %A Pham, Thi Ngan %A Phan, Thi Thom %A Nguyen, Phuoc Thao %A Ha, Quang Thuy %A VNU-UET, %B 2013 The 5th International Conference on Computational Collective Intelligence Technologies and Applications (ICCCI 2013) %C Romania %D 2013 %F SisLab:159 %T Hidden Topic Models for Multi-Label Review Classification: An Experimental Study %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/159/ %0 Conference Paper %A Pham, Thi Ngan %A Phan, Thi Thom %A Nguyen, Phuoc Thao %A Ha, Quang Thuy %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1545 %T Hidden Topic Models for Multi-Label Review Classification: An Experimental Study %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1545/ %X In recent years, Multi-Label Classification (MLC) has become an important task in the field of Supervised Learning. The MLC tasks are omnipresent in real-world problems in which an instance could belong to different classes simultaneously. In this paper, we presented a method for MLC using the hidden topic method to enrich the data features and using mutual information for feature selection. Our experiments on classifying user reviews about one thousand Vietnamese hotels showed the efficacy of the proposed approach. %0 Conference Paper %A Pham, Thi Ngan %A Tran, Van Hien %A Nguyen, Tri Thanh %A Ha, Quang Thuy %B The 9th Asian Conference on Intelligent Information and Database Systems %C Kanazawa, Japan %D 2017 %F SisLab:2320 %T Exploiting Distance graph and Hidden Topic Models for Multi-label Text Classification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2320/ %X Hidden topic models, the method to automatically detect the topics which are (hidden in a text) represented by words, have been successfully in many text mining tasks including text classification. They help to get the semantics of text by abstracting the words in text into topics. Another new method for text representation is distance graph model, which has the ability of preserving the local order of words in text, thus, enhancing the text semantics. This paper proposes a method to combine both hidden topic and distance graph models for opinion mining in hotel review domain using multi-label classification approach. Experiments show the efficiency of the proposed model provides a better performance of 4% than that of the baseline. %0 Conference Paper %A Pham, Thi Ngan %A Vuong, Thi Hong %A Thai, Thi Hoai %A Tran, Mai Vu %A Ha, Quang Thuy %B ICISA: the 2016 International Conference on Information Science and Application %C Ho Chi Minh city, Vietnam %D 2016 %F SisLab:1638 %P 1147-1156 %T Sentiment Analysis and User Similarity for Social Recommender System: An Experimental Study %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1638/ %V 376 %0 Report %9 Technical Report %A Pham, Thi Quynh Trang %D 2019 %F SisLab:3729 %T Predicting chemical-induced disease relation from literature with CNN on single concatenated input %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3729/ %0 Journal Article %@ 2588-1086 %A Pham, Thi Quynh Trang %A Bui, Manh Thang %A Dang, Thanh Hai %D 2020 %F SisLab:4204 %I VNU %J VNU Journal of Science: Computer Science and Communication Engineering %N 1 %P 11-16 %T Single Concatenated Input is Better than Indenpendent Multiple-input for CNNs to Predict Chemical-induced Disease Relation from Literature %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4204/ %V 36 %X Chemical compounds (drugs) and diseases are among top searched keywords on the PubMed database of biomedical literature by biomedical researchers all over the world (according to a study in 2009). Working with PubMed is essential for researchers to get insights into drugs’ side effects (chemical-induced disease relations (CDR), which is essential for drug safety and toxicity. It is, however, a catastrophic burden for them as PubMed is a huge database of unstructured texts, growing steadily very fast (~28 millions scientific articles currently, approximately two deposited per minute). As a result, biomedical text mining has been empirically demonstrated its great implications in biomedical research communities. Biomedical text has its own distinct challenging properties, attracting much attetion from natural language processing communities. A large-scale study recently in 2018 showed that incorporating information into indenpendent multiple-input layers outperforms concatenating them into a single input layer (for biLSTM), producing better performance when compared to state-of-the-art CDR classifying models. This paper demonstrates that for a CNN it is vice-versa, in which concatenation is better for CDR classification. To this end, we develop a CNN based model with multiple input concatenated for CDR classification. Experimental results on the benchmark dataset demonstrate its outperformance over other recent state-of-the-art CDR classification models. %0 Conference Paper %A Pham, Thi Thuong %A Nguyen, Xuan Hoai %A Nguyen, Tri Thanh %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1546 %T Research and development of methods for Data Stream Mining based on Meta Heuristic, Statistics and Nonparametric Learning %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1546/ %X Develop adaptive learning algorithms for Evolving Data Streams is one of the major challenges that we face today. In this research, we propose improved methods aim at answering three main research questions includes (1) What to remember or forget?,(2) When to do the model upgrade? And (3) How to do the model upgrade? These proposed methods are based on Meta Heuristic, Statistic and Nonparametric Learning. %0 Conference Paper %A Pham, Thi Thuong %A Nguyen, Xuan Hoai %A Nguyen, Tri Thanh %B The Advanced of International Conference on Advances in Information and Communication Technology (ICTA) %C Thai Nguyen, Vietnam %D 2016 %F SisLab:1867 %T A study on fitness representation in Genetic Programming %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1867/ %0 Journal Article %@ 0305-0548 %A Pham, Tuan Anh %A Ha, Minh Hoang %A Nguyen, Xuan Hoai %A Military Logistics Research Center, Military Logistics Academy, Vietnam, %A GRD Department, VNG Corporation, Vietnam, %A University of Engineering and Technology, VNU, Vietnam, %A FPT Technology Research Institute, FPT University, Vietnam, %A IT R&D Center, Hanoi University, Vietnam, %D 2017 %F SisLab:2573 %I Elsevier %J Computers & Operations Research %P 258-278 %T Solving the multi-vehicle multi-covering tour problem %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2573/ %V 88 %X The well-known multi-vehicle covering tour problem (m-CTP) involves finding a minimum-length set of vehicle routes passing through a subset of vertices, subject to constraints on the length of each route and the number of vertices that it contains, such that each vertex not included in any route is covered. Here, a vertex is considered as covered if it lies within a given distance of at least a vertex of a route. This article introduces a generalized variant of the m-CTP that we called the multi-vehicle multi-covering Tour Problem (mm-CTP). In the mm-CTP, a vertex must be covered at least not only once but several times. Three variants of the problem are considered. The binary mm-CTP where a vertex is visited at most once, the mm-CTP without overnight where revisiting a vertex is allowed only after passing through another vertex and the mm-CTP with overnight where revisiting a vertex is permitted without any restrictions. We first propose graph transformations to convert the last two variants into the binary one and focus mostly on solving this variant. A special case of the problem is then formulated as an integer linear program and a branch-and-cut algorithm is developed. We also develop a Genetic Algorithm (GA) that provides high-quality solutions for the problem. Extensive computational results on the new problem mm-CTP as well as its other special cases show the performance of our methods. In particular, our GA outperforms the current best metaheuristics proposed for a wide class of CTP problems. %0 Journal Article %@ 1859-3534 %A Pham, Tuan Dung %A Man, Duc Chuc %A Nguyen, Thi Nhat Thanh %A Bui, Quang Hung %A Doan, Minh Chung %D 2018 %F SisLab:3130 %J Journal of Research, Development, and Application on Information and Communication Theory %T Comparison of resampling methods on different remote sensing images for Vietnam's urban classification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3130/ %X Remotely-sensed data for urban classification is very diverse in data type, acquisition time and spatial resolution. Therefore, preprocessing is needed for input data, in which the spatial resolution must be changed by different resampling methods. However, during resampling progress, data transformations have many effects on classification results. In this research, resampling methods are evaluated; as a result, mean aggregation and bicubic interpolation methods had better results for a variety of data types. Besides, the highest overall accuracy and F1 score for urban classification maps are 98.47% and 0.9842, respectively. %0 Conference Paper %A Pham, Tuan Dung %A Man, Duc Chuc %A Nguyen, Thi Nhat Thanh %A Bui, Quang Hung %A Doan, Minh Chung %B The 8th International Conference on Knowledge and Systems Engineering (KSE) %C Hanoi, Vietnam %D 2016 %F SisLab:2777 %K Remote sensing;Satellites;Sociology;Spatial resolution;Statistics;Urban areas;Vegetation mapping;GLCNMO version2;Vietnam's urban map;optimizing;urban expansion;urbanization %P 309-314 %T Optimizing GLCNMO version 2 method to detect Vietnam's urban expansion %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2777/ %X No global scale land cover classification method performs with high accuracy at local scale. This study tries to develop a classification algorithm for urban area in Vietnam. This is the first assessment of the Global Land Cover by National Mapping Organizations (GLCNMO) version 2 method for producing global urban map in 2008. An improved and optimized algorithm is then developed based on the GLCNMO method for Vietnam taking into account of local natural and social conditions. Improving method is then applied to produce urban maps of Vietnam for the years of 2008 and 2015. Accuracy assessment showed that the improved method can achieve up to 13% higher precision and 10% higher F1 measure as compared to the global GNCNMO method. Also, an increasing trend was observed in population density in urban area in the period from 2008 to 2015 in Vietnam which may correspond to fast urbanization process in the country. The cities also tend to become less green in 2015 than 2008 as indicated by comparing the Normalized Difference Vegetation Index (NDVI) between the two years. %0 Conference Paper %A Pham, Tuan Dung %A Pham, Van Ha %A Luu, Quang Thang %A Ngo, Xuan Truong %A Nguyen, Thi Nhat Thanh %A Bui, Quang Hung %B 4th International Forum on Sustainable Future in Asia / 4th NIES International Forum %C Pan Pacific Hanoi, Vietnam %D 2019 %F SisLab:3498 %T Analyzing the impacts of urban expansion on air pollution in Vietnam using the SEAP platform %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3498/ %X The relationship between urbanization and air pollution was discovered in many studies. In this study, we analyzed the impacts of urban expansion on air pollution in Vietnam using remotely-sensed data from 2004 to 2015. In this period of time, Vietnam urban square was increased from 4623 km2 in 2004 to 5094 km2 in 2015. Besides, there is a clear difference between the average PM2.5 concentration value of urban areas and non-urban areas in Vietnam, urban PM2.5 values are generally higher than in rural areas for years. In this study, we use the SEAP (big Spatial data Exploration and Analysis Platform) platform to analyze and store data. %0 Conference Paper %A Pham, Tuan Dung %A Ta, Viet Cuong %A Pham, Thi Thanh Thuy %A Le, Thanh Ha %B KSE 2020 %C Can Tho %D 2020 %F SisLab:4153 %P 61-66 %T Reducing Blocking Artifacts in CNN-Based Image Steganography by Additional Loss Functions %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4153/ %X Our work improves the encoded image quality from HiDDeN framework, an end-to-end image steganography based on deep convolution neural network. In the encoding phase of HiDDeN framework, to embed a message in a cover image, it is required to split the cover image into smaller image blocks and embed the message bits in each block in parallel. These embedded blocks are then combined to form an encoded image that has the same size as the cover image. This image reconstruction process causes artifacts that appear on the boundaries of the blocks. This can be explained by the fact that when message bits are embedded in the image blocks, the pixel-level information of each image block is unequally alternated. In order to reduce block artifacts, in this work we propose a blocking loss as an additional objective function in HiDDeN framework. This loss measures the difference between encoded images and modified versions of the cover images. The proposed method is evaluated on COCO 2014 and BOSS datasets and the experimental results show the effectiveness in reducing the block artifacts that appeared in the encoded images of HiDDeN framework. This has an important impact on increasing the invisibility or transparency of the steganography system. In addition, the experimental result on secrecy of the proposed method also indicates the same performance as the HiDDeN pipeline. %0 Conference Paper %A Pham, Tuan Nghia %A Hoang, Thi Diep %B The 13th IEEE International Conference on Knowledge and Systems Engineering (KSE 2021) %C Bangkok, Thailand & Online %D 2021 %F SisLab:4663 %T A Distributed Algorithm for the Parsimony Bootstrap Approximation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4663/ %X Accelerating phylogenetic tree reconstruction and bootstrapping is critical, especially to support the study of the evolution of dangerous viruses. In this paper, we propose the MPBoot-MPI, a distributed algorithm efficiently implementing the idea of bootstrap approximation in MPBoot for a parallel computing environment of multiple computational nodes. MPBoot-MPI employs the master-worker paradigm and divides the work in MPBoot into three phases, each with a separate strategy to distribute computing among processes. Since the bootstrap trees are not calculated independently, processes must share results throughout task execution. We propose that when arriving at a checkpoint to report their results to the master process, worker processes apply a stochastic strategy to determine whether to perform the sending, thereby reducing the effect of latency caused by the large size of the message sent. Experiments on simulation and real benchmark datasets showed that MPBoot-MPI on multiple processes obtained MP scores and bootstrap accuracy comparable to MPBoot while achieving a promising speedup ratio. We implemented the proposed method in the MPBoot-MPI program, which is publicly accessible at https://github.com/diepthihoang/mpboot/tree/mpboot-mpi-dev. %0 Conference Paper %A Pham, Van Canh %A Ha, Kim Dung %A Ngo, Quang Dung %A Vu, Quang Cao %A Hoang, Xuan Huan %B RIVF 2016 %D 2016 %F SisLab:2373 %T A new viral marketing strategy with the competition in the large-scale Online Social Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2373/ %X The problem of Influence Maximization (IM) on social networks proposed firstly by Kempe et al. (2003) has been researched and developed with many cases. However, the IM in limited time while unwanted users are restricted is still a new potential subject. In this paper, we conducted research the problem on the model of information diffusion name Locally Bounded Diffusion and tested some useful heuristic algorithms. The results of the experiment on some real datasets of social networks show that the algorithm meta-heuristic generated better output than the others. %0 Conference Paper %A Pham, Van Canh %A Hoang, Xuan Huan %A Vu, Minh Manh %B The 4th Conference on Computational Social Networks %C Beijing, China %D 2015 %F SisLab:1718 %T Preventing and detecting infiltration on Online Social Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1718/ %0 Conference Paper %A Pham, Van Canh %A Thai, Tra My %A Ha, Kim Dung %A Ngo, Quang Dung %A Hoang, Xuan Huan %B Computational Social Networks %D 2016 %F SisLab:2372 %T Time-Critical Viral Marketing Strategy with the Competition on Online Social Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2372/ %0 Book Section %A Pham, Van Ha %A Luu, Viet Hung %A Phan, Anh %A Laffly, Dominique %A Bui, Quang Hung %A Nguyen, Thi Nhat Thanh %B TORUS 2 – Toward an Open Resource Using Services %D 2020 %F SisLab:4318 %I ISTE Ltd 2020. Published by ISTE Ltd and John Wiley & Sons, Inc %P 95-161 %T Remote Sensing Products %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4318/ %0 Conference Paper %A Pham, Van Ha %A Ngo, Xuan Truong %A Laffly, Dominique %A Jourdan, Astrid %A Nguyen, Thi Nhat Thanh %B 2019 11th International Conference on Knowledge and Systems Engineering (KSE) %D 2019 %F SisLab:3799 %P 1-6 %T Evaluation of Maximum Likelihood Estimation and regression methods for fusion of multiple satellite Aerosol Optical Depth data over Vietnam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3799/ %X This paper applied different data fusion methods including Maximum Likelihood Estimation (MLE) and Linear Regression methods on satellite images over Vietnam areas from Moderate Resolution Imaging Spectroradiometer (MODIS) and Visible Infrared Imaging Radiometer Suite (VIIRS) sensors. In comparison with ground station Aerosol Robotic Network (AERONET), the regression method is better than Maximum Likelihood Estimator (MLE). Our results show that the fusion methods can improve both data coverage and quality of satellite aerosol optical depth (AOD). Strong correlations were observed between fused AOD and AERONET AOD (R 2 = 0.8118, 0.7511 for Terra regression and MLE method, respectively). This paper presented the evaluation of data fusion algorithm and highlighted its importance on the satellite AOD data coverage and quality methods from multiple sensors. %0 Conference Paper %A Pham, Van Ha %A Nguyen, Thi Nhat Thanh %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1547 %T Mô hình hóa nồng độ bụi sử dụng ảnh vệ tinh %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1547/ %X Việc sử dụng dữ liệu ảnh sol khí trong giám sát ô nhiễm bụi là một hướng tiếp cận mới và hứa hẹn [1, 3, 5 - 8]. Các ảnh sol khí có thể được tích hợp với dữ liệu quan trắc bụi trong các mô hình ước tính mức độ ô nhiễm nhằm tăng chất lượng tính toán và dự đoán của các mô hình này. Tại Việt Nam, việc kiểm soát và tính toán lượng bụi và chất lượng không khí từ các tham số đo đạc trên các vùng thí điểm cũng đã được đề cập đến trong nhiều nghiên cứu. Tuy nhiên, việc sử dụng và khai thác các thông tin từ vệ tinh cho bài toán mô hình hoá hoặc ước lượng mức độ ô nhiễm không khí chưa được quan tâm và nghiên cứu rộng rãi tại Việt Nam. Đề tài “Mô hình hóa nồng độ bụi sử dụng ảnh vệ tinh” được nghiên cứu nhằm kế thừa và áp dụng phương pháp sử dụng ảnh vệ tinh cho bài toán Ô nhiễm không khí còn rất hạn chế ở Việt Nam. Đề tài tập trung vào giải quyết các nội dung chính như sau: ➢ Nội dung 1: Các kỹ thuật nâng cao chất lượng sản phẩm sol khí: Các phương pháp, kỹ thuật như Merging (SAA, MLE, WPC), Nội suy IO, Kriging, Phương pháp biến đổ 3D-Var, 4D-Var, bộ lọc Kalman được sử dụng nhằm nâng cao chất lượng sản phẩm sol khí. ➢ Nội dung 2: Xây dựng mô hình ước tính nồng độ Ô nhiễm bụi từ ảnh vệ tinh: Các mô hình ước tính như hồi quy, SVM, Neural Network… được thực nghiệm và đánh giá để tìm ra mô hình ước tính tốt nhất cho dữ liệu trên vùng nghiên cứu. ➢ Nội dung 3: Phân tích và đánh giá ảnh hưởng của Ô nhiễm bụi: Bản đồ nồng độ bụi PM được ước tính từ mô hình được sử dụng cho các bài toán đánh giá ảnh hưởng của Ô nhiễm không khí đế các lĩnh vực khác như sức khỏe cộng đồng [2, 4, 9] (Tỉ lệ tử vong, số lượng bệnh nhân tim mạch, ưng thư…), ảnh hưởng đến nông nghiệp, biến đối khí hậu. %0 Conference Paper %A Pham, Van Ha %A Nguyen, Thi Nhat Thanh %A Bui, Quang Hung %A Klein, Pascal %A Astrid, Jourdan %A Dominique, Laffly %B 2018 the 10th International Conference on Knowledge and Systems Engineering (KSE) %C Ho Chi Minh city, Vietnam %D 2018 %F SisLab:3132 %T Assessment of georeferencing methods on MODIS Terra/Aqua and VIIRS NPP satellite images in Vietnam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3132/ %X In this study, several georeferencing methods such as Polynomial, Thin Plate Spline (TPS) and geolocation transformations are applied on satellite images over Vietnam areas from Moderate Resolution Imaging Spectroradiometer (MODIS) sensor aboard Terra/Aqua satellites and Visible Infrared Imaging Radiometer Suite (VIIRS) sensor aboard the Suomi National Polar-orbiting Partnership (NPP) satellite. In comparison with global administrative areas data, the georeferenced images of TPS function have a better fitting than Polynomial function. In addition, the experimental results highlighted the effect of georeferencing methods on satellite aerosol optical depth (AOD) data in which TPS function produce a higher quality than Polynomial function on both MODIS Terra/Aqua and VIIRS AOD images. Validation results indicated strong correlation between AERONET AOD and TPS georeferenced images (R2 = 0.739, 0.537 and 0.596 for MODIS Terra, MODIS Aqua and VIIRS AOD, respectively) meanwhile these correlations of Polynomial are lower (R2 = 0.596, 0.443 and 0.475 for MODIS Terra, MODIS Aqua and VIIRS AOD, respectively). The correlations between MODIS AOD and VIIRS AOD images are also affected by georeferencing methods. The average of these correlations on TPS georeferenced images is 0.72 meanwhile mean correlation of Polynomial is 0.51. This study focused on the assessment of georeferencing methods and highlighted the effect of these methods on the quality and correlation of satellite AOD images. %0 Book Section %A Pham, Van Ha %A Nguyen, Thi Nhat Thanh %A Laffly, Dominique %B TORUS 2 – Toward an Open Resource Using Services %D 2020 %F SisLab:4316 %I ISTE Ltd 2020. Published by ISTE Ltd and John Wiley & Sons, Inc %P 191-224 %T Remote Sensing Case Studies %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4316/ %0 Conference Paper %A Pham, Van Huong %A Bui, Ngoc Hai %A Nguyen, Ngoc-Binh %B 2014: The 11th IEEE International Conference on Autonomic and Trusted Computing (ATC-2014) %C Bali, Indonesia %D 2014 %F SisLab:1192 %T An Approach to Instruction Scheduling at the Processor Architecture Level for Optimizing Embedded Software %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1192/ %0 Conference Paper %A Pham, Van Huong %A Nguyen, Ngoc-Binh %B 2012 International Conference on Advanced Technologies for Communications (ATC 2012) %C Hanoi, Vietnam %D 2012 %F SisLab:122 %K DSL;Embedded systems;Genetic algorithms;Linear programming;Optimization;Unified modeling language;DSL - Domain Specific Language;Embedded system;GA - Genetic Algorithm;Pareto principle;T4 - Text Template Transformation Toolkit;embedded system design;hardware-software partitioning;multi-objective optimization; %P 165 -169 %T An Approach to Design Embedded Systems by Multi-Objective Optimization %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/122/ %X Embedded system design and optimization play an important role in the development trend of embedded technology. This paper presents a new approach to design and optimize embedded systems in the design phase based on Pareto multi-objective optimization. We defined two Domain Specific Languages and developed the framework that is to design the architecture model and the component diagram of embedded systems. And we integrated the code generation technology called Text Template Transformation Toolkit to this framework to generate parameters from models automatically. Then we also do multi-objective optimization to select the best trade-off configuration of the embedded system architecture and the best hardware-software partition based on the Pareto principle and Genetic Algorithm. %0 Conference Paper %A Pham, Van Huong %A Nguyen, Ngoc-Binh %A Bui, Ngoc Hai %A Vu, Van Phuc %A VNU-UET, %B 2012: The 2012 International Conference on Integrated Circuits and Devices (ICDV 2012) %C Danang, Vietnam %D 2012 %F SisLab:1178 %P 52-57 %T Hardware-Software Co-Design to Optimize Embedded Systems by using the Pareto Principle and DSL %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1178/ %0 Conference Paper %A Pham, Van Huong %A Nguyen, Ngoc-Binh %A Vu, Van Phuc %B 2014: The 5th International Conference on Integrated Circuits, Design, and Verification (ICDV 2014) %C Hanoi, Vietnam %D 2014 %F SisLab:1193 %T A New Approach to Optimizing the Power Consumption of Existed Embedded Systems Based on the Combination of Hardware and Software %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1193/ %0 Conference Paper %A Pham, Van-Cuong %A Pham, Ngoc Hung %B 2013: 2nd International Conference on Context-Aware Systems and Applications (ICCASA 2013) %C Phu Quoc, Vietnam %D 2013 %F SisLab:769 %T A Method and Tool Support for Automated Data Flow Testing of Java Programs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/769/ %0 Conference Paper %A Pham Nghia, Luan %A Nguyen Tien, Ha %A Nguyen Van, Vinh %B 2019 16th International Conference of the Pacific Association for Computational Linguistics %D 2019 %F SisLab:3875 %T Grammatical error correction for Vietnamese using Machine Translation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3875/ %X Correction of Vietnamese grammatical errors plays an important role in Natural Language Processing. In this paper, we propose a new method using Machine Translation. We consider the grammatical error correction problem like machine translation problem with source language as grammatical wrong text and target language as grammatical right texts, respectively. Additionally, we carry out pre-processing step with grammatical wrong text using spelling checker such as MS Word spelling tool before using Machine translation model. Our experiments based on the state-of-the-art Machine Translation systems combining with pre-processing step. Experimental results achieved 84.32 BLEU score with Vietnamese grammatical error correct based on SMT architecture and 88.71 BLEU score system based on NMT architecture, which indicates that our method achieves promising results. %0 Conference Paper %A Pham Thanh, Tung %A Ma Thi, Chau %A Nguyen Manh, Tuan %A Le Dinh, Linh %A Le Thanh, Ha %B KSE-2020 %D 2020 %F SisLab:4143 %T Compression Artifacts Image Patch database for Perceptual Quality Assessment %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4143/ %X Ground truth is one of the most important component for training, testing, and benchmarking algorithms for objective quality assessment. In this paper, we propose an image patch quality database with compression artifacts. We create a new database of image patches with High Efficiency Video Coding (HEVC) compression artifacts. Then, the subjective test is conducted in a controlled environment to obtain the ground truth of image patch quality, where we collect differential mean opinion scores (DMOS) from a larger amount of observers. Finally, the rank order correlation factors between DMOS and a set of popular image quality metrics are calculated and presented. The proposed database is expected for learning patch based IQA model for block size in video rate-distortion optimization. %0 Report %9 Technical Report %A Pham Van, Long %A Vu Ngoc, Sang %A Nguyen Van, Vinh %A VNU University of Engineering and Technology, %A VNU University of Engineering and Technology, %A VNU University of Engineering and Technology, %D 2018 %F SisLab:3349 %T Study of Information Extraction in Resume %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3349/ %X This paper deals with the parsing application developed for the resumes (or CV) received in multiple formats like doc, docx, pdf, txt. These resumes can be automatically retrieved and processed by a resume information extraction system. Extracted information such as name, phone / mobile number, e-mail id, qualification, experience, skill sets etc., can be stored as a structured information in a database and then can be used in many different areas. Our system consists of 4 phases: Text Segmentation, Name Entity Recognition using Rule-based, Find Name Entities using Deep Neural Network and Text Normalization. Our work is conducted on a medium-sized collections of CV files in Vietnamese. We archived promising results with over 81% F1 for NER and also compared our model with other systems %0 Conference Paper %A Phan, Anh %A Chu, Thua Vu %A Bui, Quang Hung %A Nguyen, Thi Nhat Thanh %A Nguyen, Viet Ha %B 2018 The 5th NAFOSTED Conference on Information and Computer Science (NICS) %C Ho Chi Minh city, Vietnam %D 2018 %F SisLab:3131 %T Preliminary Result of 3D City Modelling For Hanoi, Vietnam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3131/ %X Hanoi is the one of the fastest-growing cities in Vietnam, which sets the target to turn into a smart city in 2030. Nowadays, 3D city models are being increasingly employed for many domains and tasks beyond visualization, then it will take an important role in smart city. In order to develop 3D city models, 2D geographic data such as building footprint and building height attribute are required. However, the lack of the height attribute for various types of building and low performance of rendering and visualizing 3D city models are two big remaining problems. In this paper, available data from open sources is used to predict the building height. The prediction has carried out with machine learning techniques using the combination of different attributes. After that, the models will be created using 3D tiles specification to improve the visualization performance. The preliminary results of the proposed method highlight the potential of generation of massive 3D city models from the available data in Vietnam. %0 Conference Paper %A Phan, Anh %A Man, Duc Chuc %A Bui, Quang Hung %A Nguyen, Thi Nhat Thanh %B The 9th International Conference on Knowledge and Systems Engineering (KSE 2017), 19-21 October 2017, Hue, Vietnam %D 2017 %F SisLab:2769 %T Development of Virtual Campus Using 3D GIS Technology: a case study for Vietnam National University, Hanoi. %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2769/ %0 Report %9 Technical Report %@ UET-FIT-2018xyz %A Phan, Huy Kinh %A Dang, Thanh Hai %D 2018 %F SisLab:3201 %T Protein structure prediction using Deep Learning %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3201/ %0 Conference Paper %A Phan, Xuan Hieu %A Horiguchi, Susumu %A Nguyen, Le Minh %A Nguyen, Cam Tu %B The 10th Conference of the Pacific Association for Computational Linguistics (PACLING) %C Melbourne, Australia %D 2007 %F SisLab:2301 %P 137-144 %T Semantic Analysis of Entity Contexts towards Open Named Entity Classification on the Web %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2301/ %0 Journal Article %@ 1041-4347 %A Phan, Xuan Hieu %A Nguyen, Cam Tu %A Le, Dieu Thu %A Nguyen, Le Minh %A Horiguchi, Susumu %A Ha, Quang Thuy %D 2011 %F SisLab:1888 %J IEEE Transactions on Knowledge and Data Engineering %N 7 %P 961-976 %T A Hidden Topic-Based Framework toward Building Applications with Short Web Documents %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1888/ %V 23 %X This paper introduces a hidden topic-based framework for processing short and sparse documents (e.g., search result snippets, product descriptions, book/movie summaries, and advertising messages) on the Web. The framework focuses on solving two main challenges posed by these kinds of documents: 1) data sparseness and 2) synonyms/homonyms. The former leads to the lack of shared words and contexts among documents while the latter are big linguistic obstacles in natural language processing (NLP) and information retrieval (IR). The underlying idea of the framework is that common hidden topics discovered from large external data sets (universal data sets), when included, can make short documents less sparse and more topic-oriented. Furthermore, hidden topics from universal data sets help handle unseen data better. The proposed framework can also be applied for different natural languages and data domains. We carefully evaluated the framework by carrying out two experiments for two important online applications (Web search result classification and matching/ranking for contextual advertising) with large-scale universal data sets and we achieved significant results. %0 Conference Paper %A Phan, Xuan Hieu %A Nguyen, Le Minh %A Horiguchi, Susumu %B The 17th International World Wide Web Conference (WWW) %C Beijing, China %D 2008 %F SisLab:2297 %P 91-100 %T Learning to classify short and sparse text & web with hidden topics from large-scale data collections %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2297/ %0 Journal Article %@ 0916-8532 %A Phan, Xuan Hieu %A Nguyen, Le Minh %A Inoguchi, Yasushi %A Horiguchi, Susumu %D 2007 %F SisLab:2303 %I IEICE %J IEICE Transactions on Information and Systems %N 1 %P 13-21 %T High-Performance Training of Conditional Random Fields for Large-scale Applications of Labeling Sequence Data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2303/ %V E90-D %0 Conference Paper %A Phi, Cong Huy %A Stuart, Perry %A Hoang, Van Xiem %B IEEE International Symposium on Broadband Multimedia Systems and Broadcasting 2019 %D 2019 %F SisLab:3452 %P 1-5 %T A LOW COMPLEXITY WYNER-ZIV CODING SOLUTION FOR LIGHT FIELD IMAGE TRANSMISSION AND STORAGE %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3452/ %X Compressing Light Field (LF) imaging data is a challenging but very important task for both LF image transmission and storage applications. In this paper, we propose a novel coding solution for LF images using the well-known Wyner-Ziv information theorem. First, the LF image is decomposed into 4D LF data format. Using a spiral scanning mechanismprocedure, a pseudo-sequence of LF 4D images is generated. This sequence is then compressed in a distributed coding manner as specified in the Wyner-Ziv theorem. In this context, low computational complexity can be achieved at the encoder since the high computational complexity motion estimation part is shifted to the decoderá commonly. In addition, we introduce a novel adaptive frame skipping algorithm to further explore the high correlation between 4D LF images. Experimental results show that the proposed WZ coding based LF image is able to achieve a significant compression gain, notably around 54% bitrate saving when compared with the standard High Efficiency Video Coding (HEVC) Intra benchmark while requiring less computational complexity. %0 Journal Article %@ 1568-4946 %A Phung, Manh Duong %A Ha, Quang %D 2020 %F SisLab:4075 %I Elsevier %J Applied Soft Computing %T Motion-Encoded Particle Swarm Optimization for Moving Target Search Using UAVs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4075/ %X This paper presents a novel algorithm named the motion-encoded particle swarm optimization (MPSO) for finding a moving target with unmanned aerial vehicles (UAVs). From the Bayesian theory, the search problem can be converted to the optimization of a cost function that represents the probability of detecting the target. Here, the proposed MPSO is developed to solve that problem by encoding the search trajectory as a series of UAV motion paths evolving over the generation of particles in a PSO algorithm. This motion-encoded approach allows for preserving important properties of the swarm including the cognitive and social coherence, and thus resulting in better solutions. Results from extensive simulations with existing methods show that the proposed MPSO improves the detection performance by 24\% and time performance by 4.71 times compared to the original PSO, and moreover, also outperforms other state-of-the-art metaheuristic optimization algorithms including the artificial bee colony (ABC), ant colony optimization (ACO), genetic algorithm (GA), differential evolution (DE), and tree-seed algorithm (TSA) in most search scenarios. Experiments have been conducted with real UAVs in searching for a dynamic target in different scenarios to demonstrate MPSO merits in a practical application. %0 Journal Article %@ 1568-4946 %A Phung, Manh Duong %A Quang, Ha %D 2021 %F SisLab:4436 %I Elsevier %J Applied Soft Computing %P 107376-107376 %T Safety-enhanced UAV Path Planning with Spherical Vector-based Particle Swarm Optimization %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4436/ %V 107 %X This paper presents a new algorithm named spherical vector-based particle swarm optimization (SPSO) to deal with the problem of path planning for unmanned aerial vehicles (UAVs) in complicated environments subjected to multiple threats. A cost function is first formulated to convert the path planning into an optimization problem that incorporates requirements and constraints for the feasible and safe operation of the UAV. SPSO is then used to find the optimal path that minimizes the cost function by efficiently searching the configuration space of the UAV via the correspondence between the particle position and the speed, turn angle and climb/dive angle of the UAV. To evaluate the performance of SPSO, eight benchmarking scenarios have been generated from real digital elevation model maps. The results show that the proposed SPSO outperforms not only other particle swarm optimization (PSO) variants including the classic PSO, phase angle-encoded PSO and quantum-behave PSO but also other state-of-the-art metaheuristic optimization algorithms including the genetic algorithm (GA), artificial bee colony (ABC), and differential evolution (DE) in most scenarios. In addition, experiments have been conducted to demonstrate the validity of the generated paths for real UAV operations. Source code of the algorithm can be found at https://github.com/duongpm/SPSO. %0 Journal Article %@ 1859-1256 %A Phung, Van On %A Le, Viet Ha %A Nguyen, Ngoc Hoa %D 2021 %F SisLab:4467 %J Tạp chí An toàn thông tin %T Giải pháp quản lý, đánh giá rủi ro an toàn hệ thống Công nghệ Thông tin trong Chính phủ điện tử %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4467/ %0 Journal Article %A Prativadibhayankaram, Srivatsa %A Huynh, Van Luong %A Le, Thanh Ha %A Kaup, André %D 2018 %F SisLab:3297 %J Journal of Imaging %N 7 %T Compressive Online Video Background–Foreground Separation Using Multiple Prior Information and Optical Flow %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3297/ %V 4 %X In the context of video background–foreground separation, we propose a compressive online Robust Principal Component Analysis (RPCA) with optical flow that separates recursively a sequence of video frames into foreground (sparse) and background (low-rank) components. This separation method operates on a small set of measurements taken per frame, in contrast to conventional batch-based RPCA, which processes the full data. The proposed method also leverages multiple prior information by incorporating previously separated background and foreground frames in an n-ℓ1 minimization problem. Moreover, optical flow is utilized to estimate motions between the previous foreground frames and then compensate the motions to achieve higher quality prior foregrounds for improving the separation. Our method is tested on several video sequences in different scenarios for online background–foreground separation given compressive measurements. The visual and quantitative results show that the proposed method outperforms other existing methods. %0 Conference Paper %A Prativadibhayankaram, Srivatsa %A Luong, Van Huynh %A Le, Thanh Ha %A Kaup, Andre %B The 8th International Symposium on Information and Communication Technology (SoICT) %C Nha Trang %D 2017 %F SisLab:2755 %T Compressive Online Robust Principal Component Analysis with Optical Flow for Video Foreground-Background Separation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2755/ %0 Conference Paper %A Quach, Xuan Truong %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1548 %T Báo cáo tóm tắt đề tài nghiên cứu %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1548/ %X Trong một vài thập kỉ trở lại đây, số lượng các thiết bị di động không dây ngày càng trở nên phổ biến. Đi kèm với nó là các đỏi hỏi và nhu cầu sử dụng về các dịch vụ chất lượng cao như truyền thông đa phương tiện, VoIP, VIDEO ngày càng tăng lên Nhu cầu cấp phát dải tần số cho các dịch vụ mạng không dây tăng lên một cách đáng kể dẫn đến cạn kiệt nguồn tài nguyên tần số. Tuy nhiên, trong quá trình đo đạc về mức độ sử dụng tối ưu tần số đã chỉ ra rằng phần lớn các dải tần cấp phát đang bị sử dụng một cách kém hiệu quả, mà nguyên nhân bao gồm việc cách cấp phát, quản lý và khai thác kém hiệu quả . Để tận dụng một cách hiệu quả các dải tần số, một phương pháp sử dụng và quản lý các dải tần số mới được được đề xuất bởi J Mitola với tên gọi mạng vô tuyến nhận thức - Cognitive radio networks (CRNs). Giải pháp này được xem như là một bước tiên phong cho các thế hệ tiếp theo của mạng không dây. Tuy nhiên, CRNs cũng đối mặt với những thách thức. Việc đánh giá, lựa chọn và quản lý phổ tần như thế nào để phù hợp cho nhu cầu truyền hiện tại của các đối tượng trong hệ thống mạng là một trong những thách thức không nhỏ đối với CRNs. Mặt khác, với tính chất truyền thông quảng bá tự nhiên của các mạng không dây nói chung, bất kỳ người nhận nào trong phạm vi của một truyền dẫn không dây đều có khả năng nghe được thông tin truyền đi. Do đó, bảo mật thông tin là mối quan tâm then chốt trong các mạng không dây. Do có độ phức tạp và độ trễ thấp, cũng như tính khả thi ở lớp vật lý và khả năng cùng tồn tại với các cơ chế bảo mật mã hóa hiện có mà nó có thể nâng cao mức độ tổng thể về an toàn thông tin. Vì vậy, bảo mật ở lớp vật lý đã và đang thu hút được sự quan tâm nghiên cứu của các học giả trên khắp thế giới trong thời gian gần đây. Trong phạm vi mô hình mạng CRNs khi có nhiều người dùng đồng thời trong cùng băng tần thì mức công suất của mỗi người dùng cần phải được điều khiển sao cho có thể giảm nhẹ nhiễu không mong muốn, đồng thời cũng phải đáp ứng được nhu cầu bảo mật thông tin, bảo vệ chống lại sự xâm nhập hoặc rò rỉ thông tin. Vì vậy, khả năng đảm bảo bảo mật thông tin của mạng CRNs như thế nào là vấn đề được nhiều người quan tâm. Dựa trên những hiện trạng ở trên, nghiê cứu sinh đã lựa chọn đề tài “Đánh giá hiệu năng bảo mật tại tầng vật lý trong mạng không dây” nhằm nghiên cứu, đánh giá các vấn đề liên quan đến hiệu suất hoạt động và khả năng bảo mật thông tin của các mô hình mạng nêu trên, góp phần vào giải quyết các thách thức và khó khăn trong việc hiện thực hóa hệ thống mạng vô tuyến nhận thức vào thực tiễn, làm cho công nghệ mạng không dây linh hoạt, thích nghi, đạt hiệu quả cao và an toàn trong liên lạc không dây. %0 Conference Paper %A Quach, Xuan Truong %A Tran, Hung %A Uhlemann, Elisabeth %A Tran, Truc Mai %B 22nd IEEE International Conference on Emerging Technologies and factory automation %C Limassol, Cyprus %D 2017 %F SisLab:3387 %T Secrecy Performance of Cognitive Cooperative Industrial Radio Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3387/ %X Although cognitive radio networks (CRNs) were originally intended as a powerful solution to enhance spectrum utilization, it can also be used to improve reliability by avoiding interference in the 2.4 or 5 GHz band. Using multiple relay nodes in CRNs, the outage probability, i.e., the probability that the endto-end signal-to-noise ratio drops below a predefined threshold, can be reduced significantly. This implies that the probability that a message is not delivered within a specific time frame, can be kept below a required threshold, even when there are constraints on energy efficiency in terms of peak transmit power. This is particularly useful for industrial networks with realtime constraints. However, using CRNs may also reveal secret information to eavesdroppers (EAVs). Therefore, guaranteeing secure and reliable communications in CRNs is still a challenging problem. To this end, the secrecy performance of a proactive decode-and-forward relaying scheme in a cognitive cooperative radio network is investigated. More specifically, analytical as well as approximate expressions for the secrecy outage probability and probability of non-zero secrecy capacity are derived to evaluate the system performance. Numerical results show that the approximation tightly match the analytical results and simulations, and thus it can be used to provide a fast evaluation of the security and reliability of communications using a considered assignment of relay nodes in a cognitive cooperative radio network (CCRN). Consequently, our results enable to secure the communication, and increasing the reliability, availability, robustness, and maintainability of wireless industrial network, subject to various constraints from the CRN %0 Journal Article %@ ISSN: 2169-3536 %A Quach, Xuan Truong %A Tran, Hung %A Uhlemann, Elisabeth %A Tran, Truc Mai %D 2020 %F SisLab:3995 %J IEEE Access %P 18442 -18442 %T Secrecy Performance of Cooperative Cognitive Radio Networks Under Joint Secrecy Outage and Primary User Interference Constraints %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3995/ %V 8 %X In this paper, we investigate the secrecy performance of a Cooperative Cognitive Radio Network (CCRN) in the presence of an eavesdropper (EAV). The secondary users (SUs) are subject to three constraints which include peak transmit power level and interference limitation with respect to the primary user (PU) as well as secrecy outage constraints due to the EAV. Secrecy outage is achieved when the EAV cannot decode the targeted signal, but communications in the secondary network is still possible (non-zero capacity exists). Approximation expressions of the secrecy outage probability and the probability of non-zero secrecy capacity are derived to evaluate the secrecy performance. Monte Carlo simulations are provided to examine the accuracy of the derived approximation expressions. Based on this, power allocation policies for the SUs are derived, satisfying all the constraints while maximizing the secrecy performance as well as the quality of service performance of the secondary network. It can be concluded that with knowledge of the channel state information (CSI) of the EAV it is possible to calculate the optimal value for the secrecy outage threshold of the secondary user (SU) which in turn allows maximizing the secrecy performance. Most interestingly, our numerical results illustrate that the secrecy performance of the system is much improved when the parameters obtained using the CSI of the EAV are calculated optimally. Thence, the system can adjust the power allocation so that no eavesdropping occurs even without reducing quality of service (QoS) performance compared to a network without any EAV. %0 Journal Article %@ 0031-9422 %A Remmerie, Noor %A De Vijlder, Thomas %A Laukens, Kris %A Dang, Thanh Hai %A Lemiere, Filip %A Mertens, Inge %A Valkenborg, Dirk %A Blust, Ronny %A Witters, Erwin %D 2011 %F SisLab:2076 %I Pergamon %J Phytochemistry %N 10 %P 1192-1218 %T Next generation functional proteomics in non-model plants: a survey on techniques and applications for the analysis of protein complexes and post-translational modifications %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2076/ %V 72 %0 Conference Paper %A Salem, Benferhat %A Zied, Bouraoui %A Ma, Thi Chau %A Sylvain, Lagrue %A Julien, Rossit %B 9th International Conference on Agents and Artificial Intelligence (ICCART) %C Portugal %D 2017 %F SisLab:2645 %T A Polynomial Algorithm for Merging Lightweight Ontologies in Possibility Theory Under Incommensurability Assumption %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2645/ %0 Conference Paper %A Sanya, Praseuth %A Pham, Tuan Dung %A Man, Duc Chuc %A Bui, Quang Hung %A Nguyen, Thi Nhat Thanh %B 2019 6th NAFOSTED Conference on Information and Computer Science (NICS) 2019 6th NAFOSTED Conference on Information and Computer Science (NICS) %C Hanoi %D 2019 %F SisLab:3924 %T Mapping Land Cover Types in Vientiane, Laos Using Multi-Temporal Composite Landsat 8 Images %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3924/ %0 Journal Article %@ 1550-1329 %A Seetharaman, Guna %A Le, Vu Ha %D 2006 %F SisLab:1844 %J International Journal of Distributed Sensor Networks %N 2 %P 103-119 %T Video-Assisted Global Positioning in Terrain Navigation with Known Landmarks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1844/ %V 2 %0 Book %A Sieminski, Andrzej %A Kozierkiewicz,, Adrianna %A Nunez, Manuel %A Ha, Quang Thuy %B Studies in Computational Intelligence 769 %D 2018 %E Sieminski, Andrzej %E Kozierkiewicz,, Adrianna %E Nunez, Manuel %E Ha, Quang Thuy %F SisLab:3158 %I Springer %T Modern Approaches for Intelligent Information and Database Systems. %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3158/ %V 769 %0 Journal Article %@ 1424-8220 %A Ta, Viet Cuong %A Dao, Trung Kien %A Vaufreydaz, Dominique %A Castelli, Eric %D 2020 %F SisLab:3923 %I MDPI %J Sensors %N 2 %T Collaborative Smartphone-Based User Positioning in a Multiple-User Context Using Wireless Technologies %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3923/ %V 20 %X For the localization of multiple users, Bluetooth data from the smartphone is able to complement Wi‐Fi‐based methods with additional information, by providing an approximation of the relative distances between users. In practice, both positions provided by Wi‐Fi data and relative distance provided by Bluetooth data are subject to a certain degree of noise due to the uncertainty of radio propagation in complex indoor environments. In this study, we propose and evaluate two approaches, namely Non‐temporal and Temporal ones, of collaborative positioning to combine these two cohabiting technologies to improve the tracking performance. In the Non‐temporal approach, our model establishes an error observation function in a specific interval of the Bluetooth and Wi‐Fi output. It is then able to reduce the positioning error by looking for ways to minimize the error function. The Temporal approach employs an extended error model that takes into account the time component between users’ movements. For performance evaluation, several multi‐user scenarios in an indoor environment are set up. Results show that for certain scenarios, the proposed approaches attain over 40% of improvement in terms of average accuracy %0 Conference Paper %A Ta, Viet Cuong %A Dao, Trung Kien %A Vaufreydaz, Dominique %A Castelli, Eric %B 2018 International Conference on Indoor Positioning and Indoor Navigation (IPIN) %C Nantes, France %D 2018 %F SisLab:3215 %P 206-212 %T Smartphone-based user positioning in a multiple-user context with Wi-Fi and Bluetooth %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3215/ %X In a multi-user context, the Bluetooth data from the smartphone could give an approximation of the distance between users. Meanwhile, the Wi-Fi data can be used to calculate the user’s position directly. However, both the Wi-Fi-based position outputs and Bluetooth-based distances are affected by some degree of noise. In our work, we propose several approaches to combine the two types of outputs for improving the tracking accuracy in the context of collaborative positioning. The two proposed approaches attempt to build a model for measuring the errors of the Bluetooth output and Wi-Fi output. In a non-temporal approach, the model establishes the relationship in a specific interval of the Bluetooth output and Wi-Fi output. In a temporal approach, the error measurement model is expanded to include the time component between users’ movement. To evaluate the performance of the two approaches, we collected the data from several multi-user scenarios in indoor environment. The results show that the proposed approaches could reach a distance error around 3.0m for 75 percent of time, which outperforms the positioning results of the standard Wi-Fi fingerprinting model. %0 Conference Paper %A Ta, Viet Cuong %A Vu, Thi Hong Nhan %A Bui, The Duy %B International Conference on Convergence Technology %C Vietnam %D 2012 %F SisLab:2128 %T A Multi-Agent Architecture for Smart Home %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2128/ %0 Journal Article %@ ISBN 978-604-913-300-8 %A Tan, Nguyen Duy %A Viet, Nguyen Dinh %D 2014 %F SisLab:2236 %J Kỷ yếu Hội nghị Khoa học Công nghệ Quốc gia lần thứ VII về Nghiên cứu cơ bản và ứng dụng Công nghệ thông tin (FAIR'7) %P 102-111 %T DFCB: DATA FUSION AND CHAIN-BASED ROUTING PROTOCOL FOR ENERGY-EFFICIENT IN WIRELESS SENSOR NETWORK %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2236/ %V 2 %X A unique challenge for designing routing protocols in wireless sensor network (WSN) is how to use energy efficiency. To solve this problem, in this paper, we propose data fusion and chain-base routing protocol (DFCB). In DFCB designing consists of two main works. In the first work, the chain-based routing technique is used to connect sensor nodes into a chain where each node communicates only with a nearest neighbor using distance and remaining energy of nodes to decide which node will be the chain leader. In the second work, we fuse and compress one or more packets to generate a less than size resultant packet base on the Slepian-Wolf and Dempster-Shafer theory. Simulation results show that the network lifetime of our proposed protocol can be improved about 370% and 20% compared to low-energy adaptive clustering hierarchy (LEACH) and power-efficient gathering in sensor information system (PEGASIS), respectively. %0 Journal Article %A Tan, Nguyen Duy %A Viet, Nguyen Dinh %D 2014 %F SisLab:2234 %J Proceedings of the Sixth International Conference KSE 2014 %P 61-77 %T DFTBC: Data Fusion and Tree-Based Clustering Routing Protocol for Energy-Efficient in Wireless Sensor Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2234/ %V 2 %X Energy efficiency is a unique challenge for designing routing protocols in wireless sensor network (WSN) to extend the lifetime of the entire network because sensor nodes are power constrained devices. To solve this problem, in this paper, we propose a data fusion and tree-based clustering routing algorithm (DFTBC). DFTBC designing consists of two main works. In the first work, tree-based clustering routing technique is used to connect the nodes in a cluster to form a minimum spanning tree, where each node communicates only with its nearest neighbor using distance and remaining energy of nodes to decide which node will be the cluster head (CH). In the second work, we fuse one or more packets to generate a packet with a smaller size based on the Dempster-Shafer and Slepian-Wolf theory. Our simulation results show that the network lifetime with using of our proposed protocol can be improved about 350% and 35% compared to low-energy adaptive clustering hierarchy (LEACH) and power-efficient gathering in sensor information system (PEGASIS), respectively. %0 Conference Paper %A Tan, Nguyen Duy %A Viet, Nguyen Dinh %B The 2015 International Conference Advanced Technologies for Communications (ATC) %C Ho Chi Minh city, Vietnam %D 2015 %F SisLab:2231 %P 314-319 %T SCBC: Sector-Chain Based Clustering Routing Protocol for Energy Efficiency in Heterogeneous Wireless Sensor Network %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2231/ %V 2 %X One of the most important challenges in heterogeneous wireless sensor network (WSN) is to design a routing protocol that use energy efficiently to prolong the lifetime of the entire network due to the limited battery power of sensor nodes. To solve this problem, in this paper, we propose Sector-Chain Based Clustering (SCBC) Routing Protocol, in which, entire network area is divided into sectors (clusters), which balance to the number of nodes. SCBC further reduces energy dissipation of communication in network by constructing chain for each cluster with the chain leader as the cluster head (CH) or secondary cluster head (SCH) that has high residual energy and the shortest distance between candidate nodes and base station (BS) in network. In addition, SCBC improves energy efficiency by calculating the time length of each round in the steady data transmission phase for CHs or SCHs but still guarantee that it is alive in next round. Our simulation results show that the network lifetime of our proposed protocol can be improved about 70% and 20% in comparison with PEGASIS (Power-Efficient Gathering in Sensor Information System) and IEEPB (An Improved Energy- Efficient PEGASIS-Based) protocols, respectively. Index Terms—Wireless Sensor Networks, heterogeneous sensor network, energy efficiency, routing protocol, chain-based. %0 Conference Paper %A Tan, Nguyen Duy %A Viet, Nguyen Dinh %B The 2015 IEEE RIVF International Conference on Computing Communication Technologies - Research, Innovation, and Vision for the Future %C Can Tho, Vietnam %D 2015 %F SisLab:2230 %P 180-185 %T SSTBC: Sleep Scheduled and Tree-Based Clustering Routing Protocol for Energy-Efficient in Wireless Sensor Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2230/ %V 2 %X Abstract—Since sensor nodes are battery power constrained devices in wireless sensor network (WSN), so how to use the energy of sensor nodes efficiently to prolong the network lifetime is a chief challenge for designing routing protocols. To solve this problem, in this paper, we propose sleep scheduled and tree-based clustering approach routing algorithm (SSTBC) for energy-efficient in WSN. SSTBC preserves energy by turning off radio (entering sleep mode) of either impossible or unnecessary nodes, which observe almost the same information, base on their location information to remove redundant data. In addition, to further reduce energy dissipation of communication in network, we build minimum spanning tree with the root as the cluster head (CH) from active nodes in a cluster to forward data packets to base station (BS). Our simulation results show that the network lifetime with using of our proposed protocol can be improved about 250% and 23% compared to low-energy adaptive clustering hierarchy (LEACH) and power-efficient gathering in sensor information system (PEGASIS), respectively. Keywords—Wireless Sensor Networks, energy-efficient, routing protocol, tree-based clustering. %0 Conference Paper %A Thi Le, Ngoc Anh %A Ngo, Thi Oanh %A Thi Lai, Huyen Trang %A Le, Hoang Quynh %A Nguyen, Hai Chau %A Ha, Quang Thuy %B The 8th Asian Conference on Intelligent Information and Database Systems (ACIIDS) %C Da Nang, Vietnam %D 2016 %F SisLab:1636 %P 230-240 %T An Experimental Study on Cholera Modeling in Hanoi %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1636/ %V 9622 %0 Report %9 Technical Report %A To, Van Khanh %A Nguyen, Thi Van Anh %A Nguyen, Thi Thuy %C VNU-UET %D 2018 %F SisLab:3053 %I Faculty of Information Technology %T VTSE – Verification Tool based on Symbolic Execution %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3053/ %X This paper gives an overview of modern symbolic execution techniques and presents a tool VTSE (Verification Tool based on Symbolic Execution) which allows users to verify some properties of C/C++ program based on symbolic execution technique. As two inputs including program’s source code and user’s assertion, VTSE reports whether user’s assertion are always satisfied with the program. Results of experiments performed on two sets of benchmark which are Floats-cdfpl (SV-COMP) and Kratos (FBK-IRST) are relatively positive. As for the former, VTSE has a greater advantage in solving speed although not all the problems are solved. Moreover, VTSE is also able to verify problems in Kratos benchmark which have a large number lines of code with about 500 – 2000 LOC. %0 Journal Article %@ 0162-8828 %A Toh, Kar-Ann %A Tran, Quoc Long %A Srinivasan, Dipti %D 2004 %F SisLab:3267 %I IEEE %J IEEE Transactions on pattern analysis and machine intelligence %N 6 %P 740-755 %T Benchmarking a reduced multivariate polynomial pattern classifier %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3267/ %V 26 %0 Book Section %A Toh, Kar-Ann %A Tran, Quoc Long %A Srinivasan, Dipti %B Trends in Neural Computation %D 2007 %F SisLab:3276 %I Springer %P 1-33 %T Hyperbolic function networks for pattern classification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3276/ %0 Book Section %A Toh, Kar-Ann %A Tran, Quoc Long %A Yau, Wei-Yun %B Advances in Biometric Person Authentication %D 2004 %F SisLab:3274 %I Springer %P 617-628 %T Some issues pertaining to adaptive multimodal biometric authentication %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3274/ %0 Journal Article %@ 1793-7027 %A Tran, Dang Hien %A Do, Van Tuan %A Pham, Van At %A Le, Hung Son %D 2015 %F SisLab:1471 %I World Scientific %J Journal of New Mathematics and Natural Computation %N 2 %P 121-133 %T Novel Algorithm for Nonnegative Matrix Factorization %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1471/ %V 11 %X Non-negative matrix factorization (NMF) is an emerging technique with a wide spectrum of potential applications in data analysis. Mathematically, NMF can be formulated as a minimization problem with non-negative constraints. This problem attracts much attention from researchers for theoretical reasons and for potential applications. Currently, the most popular approach to solve NMF is the multiplicative update algorithm proposed by Lee and Seung. In this paper, we propose an additive update algorithm that has a faster computational speed than Lee and Seung's multiplicative update algorithm. %0 Report %9 Technical Report %@ UET-FIT-2018xyz %A Tran, Dinh Duong %A Nguyen, Manh Hung %A Pham, Ngoc Hung %D 2018 %F SisLab:3204 %T A method of Automated User Interface Testing for Windows-based Applications %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3204/ %0 Conference Paper %A Tran, Dinh Duong %A Ogata, Kazuhiro %B International Conference on Software Engineering and Knowledge Engineering %C Wyndham Pittsburgh University Center, Pittsburgh, USA %D 2020 %F SisLab:4037 %T Formal verification of an abstract version of Anderson protocol with CafeOBJ, CiMPA and CiMPG %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4037/ %X Anderson protocol is a mutual exclusion protocol. It uses a finite Boolean array shared by all processes and the modulo (or remainder) operation of natural numbers. This is why it is challenging to formally verify that the protocol enjoys the mutual exclusion property in a sense of theorem proving. Then, we make an abstract version of the protocol called A-Anderson protocol that uses an infinite Boolean array instead. We describe how to formally specify A-Anderson protocol in CafeOBJ, an algebraic specification language and how to formally verify that the protocol enjoys the mutual exclusion property in three ways: (1) by writing proof scores in CafeOBJ, (2) with a proof assistant CiMPA for CafeOBJ and (3) with a proof generator CiMPG for CafeOBJ. We mention how to formally verify that Anderson protocol enjoys the property by showing that A-Anderson protocol simulates Anderson protocol. %0 Conference Paper %A Tran, Dinh Duong %A Ogata, Kazuhiro %A Bui, Duy Dang %A Gupta, Parth %B THE 27TH ASIA-PACIFIC SOFTWARE ENGINEERING CONFERENCE %D 2020 %F SisLab:4284 %T Lemma Weakening for State Machine Invariant Proofs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4284/ %X Lemma conjecture is one of the most challenging tasks in theorem proving. The paper focuses on invariant properties (or invariants) of state machines. Thus, lemmas are also invariants. To prove that a state predicate p is an invariant of a state machine M, in general, we need to find an inductive invariant q of M such that q(s) implies p(s) for all states s of M. q is often in the form p∧p', and p' is often in the form q1 ∧...∧qn. q1, ..., qn are the lemmas of the proof that p is an invariant of M. The paper proposes a technique called Lemma Weakening (LW). LW replaces qi with qi' such that qi(s) implies qi'(s) for all states s of M, which can make the proof reasonably tractable that may become otherwise unreasonably hard. MCS mutual exclusion protocol is used as an example to demonstrate the power of LW. %0 Conference Paper %A Tran, Dinh Duong %A Pham, Ngoc Hung %A Nguyen, Duy Tung %A Toshiba Software Development Vietnam, %B 2018 the 9th International Symposium on Information and Communication Technology (SoICT) %C Da Nang City, Viet Nam %D 2018 %F SisLab:3124 %T A Method for Automated User Interface Testing of Windows-based Applications %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3124/ %0 Journal Article %@ 2079-9292 %A Tran, Dinh Lam %A Tran, Xuan Tu %A Bui, Duy Hieu %A Pham, Cong Kha %D 2020 %F SisLab:3951 %I MDPI %J Electronics %N 4 %P 684 %T An Efficient Hardware Implementation of Residual Data Binarization in HEVC CABAC Encoder %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3951/ %V 9 %X HEVC-standardized encoders employ the CABAC (context-based adaptive binary arithmetic coding) to achieve high compression ratios and video quality that supports modern real-time high-quality video services. Binarizer is one of three main blocks in a CABAC architecture, where binary symbols (bins) are generated to feed the binary arithmetic encoder (BAE). The residual video data occupied an average of 75% of the CABAC’s work-load, thus its performance will significantly contribute to the overall performance of whole CABAC design. This paper proposes an efficient hardware implementation of a binarizer for CABAC that focuses on low area cost, low power consumption while still providing enough bins for high-throughput CABAC. On the average, the proposed design can process upto 3.5 residual syntax elements (SEs) per clock cycle at the maximum frequency of 500 MHz with an area cost of 9.45 Kgates (6.41 Kgates for the binarizer core) and power consumption of 0.239 mW (0.184 mW for the binarizer core) with NanGate 45 nm technology. It shows that our proposal achieved a high overhead-efficiency of 1.293 Mbins/Kgate/mW, much better than the other related high performance designs. In addition, our design also achieved a high power-efficiency of 8288 Mbins/mW; this is important factor for handheld applications. %0 Conference Paper %A Tran, Duc Duong %A Tan, Hanh %A Pham, Son Bao %B The 8th International Conference on Knowledge and Systems Engineering (KSE) %C Hanoi, Vietnam %D 2016 %F SisLab:2258 %P 91-95 %T Customer gender prediction based on E-commerce data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2258/ %0 Report %9 Technical Report %@ UET-FIT-2018xyz %A Tran, Duc Toan %A Pham, Ngoc Hung %D 2018 %F SisLab:3203 %T "Smart Codefun": A tool supporting programming skills for students %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3203/ %0 Conference Paper %A Tran, Hoang Viet %A Le, Chi Luan %A Nguyen, Quang Trung %A Pham, Ngoc Hung %B 2014: The 6th International Conference on Knowledge and Systems Engineering (KSE 2014) %C Hanoi, Vietnam %D 2014 %F SisLab:1297 %T An Efficient Method for Automated Generating Models of Component-Based Software. %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1297/ %0 Conference Paper %A Tran, Hoang Viet %A Le, Chi Luan %A Pham, Ngoc Hung %B The 2016 IEEE RIVF International Conference on Computing and Communication Technologies %C Hanoi, Vietnam %D 2016 %F SisLab:1985 %P 1-6 %T A Strongest Assumption Generation Method for Component-Based Software Verification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1985/ %X This paper presents a method for generating strongest assumptions during component-based software verification. The key idea of this method is to improve the technique for answering membership queries of the Teacher when responding to the Learner in the %0 Conference Paper %A Tran, Hoang Viet %A Nguyen, Quang-Trung %A Pham, Ngoc Hung %B THE 10TH INTERNATIONAL SYMPOSIUM ON INFORMATION AND COMMUNICATION TECHNOLOGY %C Ha Long Bay, Quang Ninh, Vietnam %D 2019 %F SisLab:3598 %T On Implementation of the Improved Assume-Guarantee Verification Method for Timed Systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3598/ %0 Journal Article %@ 2588-1086 %A Tran, Hoang Viet %A Pham, Ngoc Hung %D 2018 %F SisLab:3500 %I VNU %J VNU Journal of Science: Computer Science and Communication Engineering %N 2 %P 16-32 %T On Locally Strongest Assumption Generation Method for Component-Based Software Verification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3500/ %V 34 %X Assume-guarantee reasoning, a well-known approach in component-based software (CBS) verification, is in fact a language containment problem whose computational cost depends on the sizes of languages of the software components under checking and the assumption to be generated. Therefore, the smaller language assumptions, the more computational cost we can reduce in software verification. Moreover, strong assumptions are more important in CBS verification in the context of software evolution because they can be reused many times in the verification process. For this reason, this paper presents a method for generating locally strongest assumptions with locally smallest languages during CBS verification. The key idea of this method is to create a variant technique for answering membership queries of the Teacher when responding to the Learner in the L–based assumption learning process. This variant technique is then integrated into an algorithm in order to generate locally strongest assumptions. These assumptions will effectively reduce the computational cost when verifying CBS, especially for large–scale and evolving ones. The correctness proof, experimental results, and some discussions about the proposed method are also presented. %0 Conference Paper %A Tran, Hoang Viet %A Pham, Ngoc Hung %A Dang, Van Hung %B 2018 10th International Conference on Knowledge and Systems Engineering (KSE) %C Ho Chi Minh city, Vietnam %D 2018 %F SisLab:3120 %T On Improvement of Assume-Guarantee Verification Method for Timed Component-Based Software %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3120/ %0 Journal Article %@ 0916-8532 %A Tran, Hoang Viet %A Pham, Ngoc Hung %A Nguyen, Viet Ha %D 2019 %F SisLab:3503 %I IEICE %J IEICE Transactions on Information and Systems %T ON LOCALLY MINIMUM AND STRONGEST ASSUMPTION GENERATION METHOD FOR COMPONENT-BASED SOFTWARE VERIFICATION %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3503/ %X Since software becomes more complex during its life cycle, the verification cost becomes higher, especially for such methods which are using model checking in general and assume-guarantee reasoning in specific. To address the problem of reducing the assume-guarantee verification cost, this paper presents a method to generate locally minimum and strongest assumptions for verification of component-based software. For this purpose, we integrate a variant of membership queries answering technique to an algorithm which considers candidate assumptions that are smaller and stronger first, larger and weaker later. Because the algorithm stops as soon as it reaches a conclusive result, the generated assumptions are the locally minimum and strongest ones. The correctness proof of the proposed algorithm is also included in the paper. An implemented tool, test data, and experimental results are presented and discussed. %0 Conference Paper %A Tran, Hoang-Viet %A Pham, Ngoc Hung %A Le, Chi-Luan %A Nguyen, Quang-Trung %B 2014: The 6th International Conference on Knowledge and Systems Engineering (KSE 2014) %C Hanoi, Vietnam %D 2014 %F SisLab:445 %P 499-511 %T An Efficient Method for Automated Generating Models of Component-Based Software %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/445/ %0 Conference Paper %A Tran, Hong Viet %A Nguyen, Van Vinh %A Nguyen, Le Minh %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1549 %T Improving English-Vietnamese Statistical Machine Translation Using Preprocessing Dependency Syntactic %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1549/ %X In this paper, we would like to present a new reordering approach based on a dependency parser in phrase based statistical machine translation (SMT) for English to Vietnamese. The proposed method can efficiently incorporate linguistic knowledge into SMT systems. We inspired from [1] using preprocessing reordering approaches. Dependency parser and transformation rules are used to reorder the source sentence and applied for systems translating English to Vietnamese. The experiment results showed that the proposed approach achieved improvements in BLEU scores over MOSES which is the state-of-the art phrase based SMT system. %0 Conference Paper %A Tran, Hong Viet %A Nguyen, Van Vinh %A Nguyen, Le Minh %B The 2015 Conference of the Pacific Association for Computational Linguistics %C Bali, Indonesia %D 2015 %F SisLab:1716 %T Improving English-Vietnamese Statistical Machine Translation Using Preprocessing Dependency Syntactic %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1716/ %0 Conference Paper %A Tran, Hung Quan %A Nguyen, Dinh Nien %A Do, Duc Kien %A Nguyen, Khanh Thinh %A Tran, Hai Dang %A Nguyen, LeMinh %A Pham, Bao Son %B 2014: The 6th International Conference on Knowledge and Systems Engineering (KSE 2014) %C Hanoi, Vietnam %D 2014 %F SisLab:439 %P 117-128 %T A Community-Based Vietnamese Question Answering System %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/439/ %0 Conference Paper %A Tran, Huu Quoc Dong %A Chu, Duc Trinh %A Bui, Thanh Tung %A Nguyen, Thi Thanh Van %A Phan, Hoang Anh %A Nguyen, Dinh Tuan %A Bui, Duy Nam %B The 5 th International Conference on Engineering Mechanics and Automation (ICEMA 5) Ha Noi %C Ha Noi %D 2019 %F SisLab:3764 %T Employing Extended Kalman Filter with Indoor Positioning System for Robot Localization Application %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3764/ %X IPS (Indoor Positioning System) is used to localize the robot position in a narrow environment. However, IPS signal is not accurate in some cases, the IPS signal transmission will be obstructed by people the public museums, for example. The combination with other sensors will reduce the poor signal from IPS and noise in the environment with many obstacles. The Extended Kalman Filter (EKF) have been widely used for mobile robot localization system and gained certain results. In this paper, EKF is embedded to Central Processing Unit with Robot Operating System on “UET-FuSo” robot to fuse IPS signal with encoder and IMU sensor for the determination of position and orientation robot. The obtained results show the effect of the proposed method when the robot localization is more stable and accurate than using only IPS signal. This localization will be applied for mapping and navigation of mobile robot in exhibition guidance robot. %0 Conference Paper %A Tran, Huy Hoang %A Nguyen, The Hoang Anh %A Do, Quoc Vuong %A Le, Thanh Ha %B Một số vấn đề chọn lọc của Công nghệ thông tin và Truyền thông 2018 %D 2018 %F SisLab:3292 %T EOG Detection using Haar Wavelet Transform toward implementation of an IOT Brain computer interface %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3292/ %0 Report %9 Technical Report %A Tran, Mai Vu %D 2019 %F SisLab:3716 %T Hệ thống tóm tắt sự kiện %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3716/ %X Tóm tắt văn bản là một vấn đề nhận được nhiều sự quan tâm của cộng đồng nghiên cứu với rất nhiều các bài báo khoa học xuất hiện tại các hội nghị lớn như: ACL, DUC, TAC, COLING, EMNLP... Mục đích chính của tóm tắt văn bản đó chính là đưa ra văn bản ngắn gọn và xúc tích hơn từ một văn bản đầu vào. Bài toán này cũng được ứng dụng nhiều trong các hệ thống thực tế như: tóm tắt các trang web trong máy tìm kiếm, tóm tắt các tin tức, tóm tắt các quan điểm người dùng… Một trong những bài toán đang được quan tâm hiện nay đối với tóm tắt đó chính là tóm tắt sự kiện. %0 Conference Paper %A Tran, Mai Vu %A Le, Hoang Quynh %A Phi, Van Thuy %A Pham, Thanh Binh %A Nigel, Collier %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1550 %T Exploring a Probabilistic Earley Parser for Event Composition in Biomedical Texts %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1550/ %X We describe a high precision system for extracting events of biomedical significance that was developed during the BioNLP shared task 2013 and tested on the Cancer Genetics data set. Our system explored a multi-stage approach including trigger detection, edge detection and event composition. After trigger edge detection is finished we are left with a semantic graph from which we must select the optimal subset that is consistent with the semantic frames for each event type. The system achieved an F-score on the development data of 73.67 but was ranked 5th out of six with an F-score of 29.94 on the test data. How-ever, precision was the second highest ranked on the task at 62.73. Analysis suggests the need to continue to improve our system for complex events particularly taking into account cross-domain differences in argument distributions. %0 Conference Paper %A Tran, Mai Vu %A Nguyen, Minh Hoang %A Nguyen, Sy Quan %A Nguyen, Minh Tien %A Phan, Xuan Hieu %A VNU-UET, %B 2012 Fourth International Conference on Knowledge and Systems Engineering (KSE) %C Danang, Vietnam %D 2012 %F SisLab:100 %K Data mining;Detectors;Entropy;Feature extraction;Fires;Machine learning;Monitoring;Internet;information retrieval;learning (artificial intelligence);natural language processing;text analysis;F1 measure;Vietnamese language;Vietnamese news;VnLoc;Web data;crime;decide support tools;event extraction method;event online monitoring system;fire;free text;information extraction;lexico-semantic learning;machine learning;precision;real-time news event extraction framework;recall;risk analysis system;transport accident; %P 161 -166 %T VnLoc: A Real — Time News Event Extraction Framework for Vietnamese %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/100/ %X Event Extraction is a complex and interesting topic in Information Extraction that includes event extraction methods from free text or web data. The result of event extraction systems can be used in several fields such as risk analysis systems, online monitoring systems or decide support tools. In this paper, we introduce a method that combines lexico — semantic and machine learning to extract event from Vietnamese news. Furthermore, we concentrate to describe event online monitoring system named VnLoc based on the method that was proposed above to extract event in Vietnamese language. Besides, in experiment phase, we have evaluated this method based on precision, recall and F1 measure. At this time of experiment, we on investigated on three types of event: FIRE, CRIME and TRANSPORT ACCIDENT. %0 Conference Paper %A Tran, Mai-Vu %A Collier, Nigel %A Le, Hoang-Quynh %A Phi, Van-Thuy %A Pham, Thanh-Binh %B 2013 the BioNLP Shared Task 2013 Workshop. Association for Computational Linguistics (ACL) %C Bulgaria %D 2013 %F SisLab:192 %T Exploring a Probabilistic Earley Parser for Event Decomposition in Biomedical Texts %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/192/ %0 Conference Paper %A Tran, Manh Cuong %A Tran, Trung Kien %A Nguyen, Minh Tan %A Nguyen, Thu Trang %A Nguyen, Son %A Vo, Dinh Hieu %B THE 28TH ASIA-PACIFIC SOFTWARE ENGINEERING CONFERENCE %D 2021 %F SisLab:4686 %T API parameter recommendation based on language model and program analysis %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4686/ %X APIs are extensively and frequently used in source code to leverage existing libraries and improve programming productivity. However, correctly and effectively using APIs, es- pecially from unfamiliar libraries, is a non-trivial task. Although various approaches have been proposed for recommending API method calls in code completion, suggesting actual parameters for such APIs still needs further investigating. In this paper, we introduce FLUTE, an efficient and novel approach combining program analysis and language models for recommending API parameters. With FLUTE, the source code of programs is first analyzed to generate syntactically legal and type-valid candidates. Then, these candidates are ranked using language models. Our empirical results on two large real-world projects Netbeans and Eclipse indicate that FLUTE achieves 80% and +90% in Top- 1 and Top-5 Precision, which means the tool outperforms the state-of-the-art approach. %0 Journal Article %@ 1939-0114 %A Tran, Nghi Phu %A Hoang, Dang Kien %A Ngo, Quoc Dung %A Nguyen, Dai Tho %A Nguyen, Ngoc Binh %A VNU University of Engineering and Technology, %A People’s Security Academy, %A Posts and Telecommunications Institute of Technology, %A Kyoto College of Graduate Studies for Informatics, %D 2019 %F SisLab:3757 %I Hindawi %J Security and Communication Networks %T A Novel Framework to Classify Malware in MIPS Architecture-based IoT Devices %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3757/ %X Malware on devices connected to the Internet via the Internet of Things (IoT) ) is evolving and is a core component of the fourth industrial revolution. IoT devices use the MIPS architecture with a large proportion running on embedded Linux operating systems, but the automatic analysis of IoT malware has not resolved. We proposed a framework to classify malware in IoT devices by using MIPS-based system behavior (system call - syscall) got from our F-Sandbox passive process and machine learning techniques. The F-Sandbox is a new type for IoT sandbox, automatically created from the real firmware of the specialized IoT devices, inheriting the specialized environment in the real firmware, therefore creating a diverse environment for sandboxing as an important characteristic of IoT sandbox. This framework classifies five families of IoT malware with F1-Weight = 97.44%. %0 Conference Paper %A Tran, Nghi Phu %A Le, Huy Hoang %A Nguyen, Ngoc Toan %A Nguyen, Dai Tho %A Nguyen, Ngoc Binh %A VNU University of Engineering and Technology, %A People’s Security Academy, %B 19th International Symposium on Communications and Information Technologies (ISCIT 2019) %C Ho Chi Minh City %D 2019 %F SisLab:3765 %P 568-573 %T C500-CFG: A Novel Algorithm to Extract Control Flow-Based Features for IoT Malware Detection %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3765/ %X Control flow-based features proposed by Ding, static characteristic extraction method, has the ability to detect malicious code with higher accuracy than traditional Text-based methods. However, this method resolved NP-hard problem in a graph, therefore it is not feasible with the large-size and highcomplexity programs. So, we propose the C500-CFG algorithm in Control flow-based features based on the idea of dynamic programming, solving Ding’s NP-hard problem by polynomial complexity O(N^2) algorithm, where N is the number of basic blocks in decompiled executable codes. Our algorithm is more efficient and more outstanding in detecting malware than Ding’s algorithm: fast processing time, allowing processing large files, using less memory and extracting more feature information. Applying our algorithms with IoT data sets gives outstanding results on 2 measures: Accuracy = 99.34%, F1-Score = 99.32%. %0 Conference Paper %A Tran, Nghi Phu %A Le, Huy Hoang %A Nguyen, Ngoc Toan %A Nguyen, Dai Tho %A Nguyen, Ngoc Binh %A VNU University of Engineering and Technology, %A People’s Security Academy, %A Kyoto College of Graduate Studies for Informatics, %B 10th International Symposium on Information and Communication Technology (SoICT 2019) %C Ha Noi - Ha Long %D 2019 %F SisLab:3763 %P 248-254 %T CFDVex: A Novel Feature Extraction Method for Detecting Cross-Architecture IoT Malware %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3763/ %X The widespread adoption of Internet of Things (IoT) devices built on different architectures gave rise to the creation and development of multi-architecture malware for mass compromise. Crossarchitecture malware detection plays an important role in detecting malware early on devices using new or strange architectures. Prior knowledge of malware detection on traditional architectures can be inherited for the same task on new and uncommon ones. Basing on CFD and Vex intermediate representation, we propose a feature selection method to detect cross-architecture malware, called CFDVex. Experimental evaluation of the proposed approach on our large IoT dataset achieved good results for cross-architecture malware detection. We only trained a SVM model by Intel 80386 architecture samples, our method could detect the IoT malware for the MIPS architecture samples with 95.72% of accuracy and 2.81% false positive rate. %0 Conference Paper %A Tran, Nghi Phu %A Ngo, Quoc Dung %A Hoang, Dang Kien %A Nguyen, Ngoc Binh %A Nguyen, Dai Tho %A Academy of Popular Security, %A University of Engineering and Technology, Vietnam National University, Hanoi, %B Hội thảo lần thứ III Một số vấn đề chọn lọc về an toàn an ninh thông tin %C Da Nang %D 2018 %F SisLab:3379 %T Phát hiện mã độc trên các thiết bị IoT dựa trên lời gọi Syscall và phân loại một lớp SVM %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3379/ %X - Mã độc trên các thiết bị Vạn vật kết nối Internet - Internet of Things (IoT), thành phần cốt lõi trong cuộc cách mạng 4.0, đang ngày càng tăng nhanh. Các thiết bị IoT có kiến trúc MIPS chiếm tỉ lệ lớn, song các nghiên cứu phát hiện mã độc dựa trên hành vi của các thiết bị này chưa được đề cập. Chúng tôi đề xuất quy trình phân tích phát hiện mã độc trong các thiết bị IoT sử dụng kiến trúc MIPS dựa trên lời gọi hệ thống syscall (hay system call) bằng kỹ thuật phân loại một lớp SVM. Các chương trình được chạy trong C500-Sandbox để thu thập các syscall được gọi, các syscall được biểu diễn dưới dạng đặc trưng n-gram, sau đó sử dụng phương pháp phân tích thành phần chính (Principal Component Analysis - PCA) để giảm số chiều trước khi được huấn luyện/nhận dạng bằng mô hình phân loại một lớp SVM (One class SVM). Quy trình cho kết quả phát hiện tốt với chỉ số F-Score (F1) = 0,976, Average precision (AP) = 0,992, Accuracy (AC) = 0,956. %0 Journal Article %@ 2278-3075 %A Tran, Nghi Phu %A Ngo, Quoc Dung %A Le, Van Hoang %A Nguyen, Dai Tho %A Nguyen, Ngoc Binh %A VNU University of Engineering and Technology, %A People’s Security Academy, %D 2019 %F SisLab:3761 %I Blue Eyes Intelligence Engineering & Sciences Publication %J International Journal of Innovative Technology and Exploring Engineering (IJITEE) %N 11 %P 32-40 %T A System Emulation for Malware Detection in Routers %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3761/ %V 8 %X Nowadays, there are many discussions on the fourth industrial revolution with a combination of real physical and virtual systems (Cyber Physical Systems), Internet of Things (IoT) and Internet of Services (IoS). Along with this revolution is the rapid development of malicious code on IoT devices, leading to not only the risk of personal privacy information leaking but also the risk of network security in general. In this paper, we propose C500-toolkit, a novel tool for malware detection in Commercial-off-the-shelf routers, based on dynamic analysis approach. The main contribution of C500-toolkit is to provide an environment for fully emulating router firmware image including both operating system and web-interface. To show the advantage of C500-toolkit, experiments of this tool with embedded malwares Linux/TheMoon and Linux/Mirai are presented. %0 Journal Article %@ 0010-4620 %A Tran, Nghi Phu %A Nguyen, Dai Tho %A Le, Huy Hoang %A Nguyen, Ngoc Toan %A Nguyen, Ngoc Binh %A People's Security Academy, %A VNU University of Engineering and Technology, %A The Kyoto College of Graduate Studies for Informatics, %D 2020 %F SisLab:3998 %I Oxford University Press %J The Computer Journal %T An Efficient Algorithm to Extract Control Flow-based Features for IoT Malware Detection %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3998/ %X Control flow-based feature extraction method has the ability to detect malicious code with higher accuracy than traditional text-based methods. Unfortunately, this method has been encountering with the NP-hard problem, which is infeasible for the large-sized and high-complexity programs. To tackle this, we propose a control flow-based features extraction dynamic programming algorithm (CFD) for fast extraction of control flow-based features with polynomial time O(N^2), where N is the number of basic blocks in decompiled executable codes. From the experimental results, it is demonstrated that the proposed algorithm is more efficient and effective in detecting malware than the existing ones. Applying our algorithm to an IoT dataset gives better results on 3 measures: Accuracy (AC) = 99.05%, False Positive Rate (FPR) = 1.31% and False Negative Rate (FNR) = 0.66%. %0 Conference Paper %A Tran, Nghi Phu %A Nguyen, Huy Trung %A Ngo, Quoc Dung %A Nguyen, Ngoc Binh %A Nguyen, Dai Tho %A People's Security Academy, %A University of Engineering and Technology, Vietnam National University, Hanoi, %B Hội thảo lần thứ I Một số vấn đề chọn lọc về An toàn an ninh thông tin %C Hanoi, Vietnam %D 2016 %F SisLab:2362 %T Phát triển công cụ dịch ngược firmware trên thiết bị định tuyến %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2362/ %X Vạn vật kết nối (Internet of Things) hiện nay được sử dụng ngày càng rộng rãi trong cuộc sống, đi kèm với những tiện ích mới công nghệ này mang lại đó chính là vấn đề an ninh, an toàn thông tin. Kết quả các nghiên cứu gần đây đã chỉ ra rằng các lỗ hổng bảo mật và đặc biệt là mã độc xuất hiện rất nhiều trên hệ điều hành của các thiết bị mạng (firmware). Vì tính đặc thù cao của các thiết bị mạng cũng như của các firmware mà việc phân tích, rà quét các lổ hổng bảo mật và mã độc gặp rất nhiều trở ngại. Hơn nữa, do chưa có được sự quan tâm đúng mức của cá nhân, tổ chức lớn nên việc phát triển các công cụ cũng như phương pháp phân tích, phát hiện còn nhiều hạn chế. Hạn chế có thể kể đến là khả năng trích chọn firmware và dịch ngược chúng thành mã tường minh để từ đó có thể sử dụng các phương pháp phân tích lỗ hổng, mã độc khác nhau. Trong bài báo này, C500-Toolkit sẽ được giới thiệu với mục tiêu tăng hiệu suất trích chọn và dịch ngược firmware của các thiết bị định tuyến. Gần 13.674 firmware từ 27 nhà phân phối được sử dụng để đánh giá và so sánh với các công cụ hiện có như fmk, binwalk và mô đun trích chọn, dịch ngược của firmadyne. %0 Conference Paper %A Tran, Nghi Phu %A Nguyen, Ngoc Binh %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1551 %T Phân tích và phát hiện lỗ hổng bảo mật trong firmware của các thiết bị mạng %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1551/ %X Công nghệ ngày càng phát triển mãnh mẽ, đi sâu vào mọi khía cạnh của cuộc sống, do đó vấn đề an ninh an toàn trở thành một trong những vấn đề được quan tâm nhất trong thế giới ngày nay. Nguy cơ về mất an ninh thông tin không chỉ dừng lại mức ứng dụng phần mềm mà đặt ra cả ở mức phần cứng thiết bị và ngày nay một lĩnh vực đang dấy lên nhiều lo ngại là nguy cơ từ các hệ thống nhúng tích hợp, firmware của các thiết bị, đặc biệt các thiết bị đóng vai trò trung tâm trong kết nối, truyền dẫn thông tin như các thiết bị mạng. Các nghiên cứu từ trước tập trung cho các ứng dụng mức cao trên kiến trúc i386, hệ điều hành Windows, các phương pháp và công cụ hỗ trợ cho việc phân tích firmware còn rất hạn chế và gặp nhiều khó khăn do tiếp cận mức thấp, kiến trúc đa dạng, thiếu tài liệu đặc tả đầy đủ. Nghiên cứu của chúng tôi tập trung vào việc tìm hiểu kiến trúc của các thiết bị mạng, từ đó tìm cách trích xuất firmware từ các thiết bị này, sau đó phân tích nhằm phát hiện có hay không các lỗ hổng bảo mật hay malware cài cắm trong firmware. Việc phân tích và phát hiện dựa trên 2 phương pháp tĩnh và động. Các firmware được tiến hành phân tách, thu thập lấy phần thực thi chính của firmware thường là các Embeded Web trên nền Linux 2.6, phần lõi này được tiến hành phân tích tĩnh tìm ra backdoor hoặc chạy trong môi trường mô phỏng để quét phát hiện lỗ hổng bảo mật. %0 Conference Paper %A Tran, Ngoc Anh %A Nguyen, Phuong Thai %A Dao, Thanh Tinh %A Nguyen, Hong Quan %B 2015: The 11th IEEE-RIVF International Conference on Computing and Communication Technologies (RIVF 2015) %C Can Tho, Vietnam %D 2015 %F SisLab:1186 %T Identifying Reduplicative Words for Vietnamese Word Segmentation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1186/ %0 Conference Paper %A Tran, Ngoc Ha %A Do, Duc Dong %A Hoang, Xuan-Huan %B 2013 International Conference on Computing, Management and Telecommunications (ComManTel) %C HoChiMinh city %D 2013 %F SisLab:136 %K Algorithm design and analysis;Ant colony optimization;Greedy algorithms;Heuristic algorithms;Proteins;Runtime;Vectors;Ant Colony Optimization;Local Search;Multiple Graph Alignment;Pheromone update rule;label %P 386-391 %T An Efficient Ant Colony Optimization Algorithm for Multiple Graph Alignment %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/136/ %X The Multiple Graph Alignment (MGA) is a new method to analyze the structure of biological molecules. This method allows detect functional similarities in the structure of biological systems. This article introduces an ant colony optimization algorithm combined with local search for optimal align multi-graph analysis of protein structures. Experiment results showed that the new algorithm outperformed the other heuristic approach and existing evolutionary computing. %0 Conference Paper %A Tran, Ngoc Ha %A Do, Duc Dong %A Hoang, Xuan-Huan %B 2014: The 11th IEEE International Conference on Autonomic and Trusted Computing (ATC-2014) %C Bali, Indonesia %D 2014 %F SisLab:1188 %P 181-186 %T A Novel Ant Based Algorithm for Multiple Graph Alignment %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1188/ %0 Conference Paper %A Tran, Ngoc Ha %A Hoang, Xuan Huan %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1553 %T An Efficient Ant Based Algorithm for Global Alignment of Protein-Protein Interaction Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1553/ %X Global aligning two protein-protein interaction networks is an essentially important task in bioinformatics. It is a challenging and widely studied research topic in recent years. Accurately aligned networks allow us to identify functional modules of proteins and/ororthologous proteins from which unknown functions of a protein can be inferred. In this article, we introduce an ant based global network alignment algorithm called ACOGA. The experiments show that the method that we proposed get better results than the introduced methods recently. %0 Conference Paper %A Tran, Ngoc Ha %A Hoang, Xuan Huan %B Hội nghị khoa học quốc gia lần thứ 8 Nghiên cứu cơ bản và ứng dụng CNTT (FAIR 2015) %C Hanoi, Vietnam %D 2015 %F SisLab:1719 %T Một thuật toán tối ưu đàn kiến dóng hàng toàn cục mạng tương tác protein %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1719/ %0 Conference Paper %A Tran, Ngoc Ha %A Hoang, Xuan-Huan %B KSE 2016 %D 2016 %F SisLab:2371 %T ACOGNA: An Efficient Method for Protein-Protein Interaction Network Alignment %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2371/ %X Protein–protein interaction network alignment enables us to identify orthologous proteins, predict protein functions and build evolutionary relationships of species. This article introduces an algorithm for global alignment of protein- protein interaction network based on ant colony optimization (ACO) method. The experimental results offer outstanding advantages of the proposed algorithms. %0 Conference Paper %A Tran, Nguyen Huong %A Do, Minh Kha %A Tran, Hoang Viet %A Pham, Ngoc Hung %B 12th International Conference on Knowledge and Systems Engineering (KSE 2020) %C Can Tho, Vietnam %D 2020 %F SisLab:4069 %T Generate Test Data from C/C++ Source Code using Weighted CFG and Boundary Values %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4069/ %X This paper presents two test data automatic generation methods which are based on weighted control flow graph (named WCFT) and boundary values of input parameters (named BVTG). Firstly, WCFT method generates a CFG from a given unit function, updates weight for it, then generates test data from the greatest weight test paths. In the meantime, WCFT can find dead code that can be used for automatic source code errors fix. Secondly, BVTG method generates test data from boundary values of input parameters of the given unit function. The combination of the two generated test data sets from these two methods will improve the error detection ability while maintaining a high code coverage. An implemented tool (named WCFT4Cpp) and experimental results are also presented to show the effectiveness of the two proposed methods in both time required to generate test data and error detection ability. %0 Conference Paper %A Tran, Nguyen Le %A Do, Thanh Cong %A Nguyen, Ba Tung %A Ta, Viet Cuong %A Le, Thanh Ha %B 2013: 6th International Conference on Signal Processing, Image Processing and Pattern Recognition (SIP) %D 2013 %F SisLab:791 %P 60-65 %T Contour Based Hand Gesture Recognition Using Depth Data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/791/ %0 Report %9 Technical Report %@ TR2018-FIT-01 %A Tran, Nhu Thuat %A Nguyen, Huu Hong %A Nguyen, Thanh Tung %A Dang, Tien Son %A Luong, Thai Le %A Phan, Xuan Hieu %D 2018 %F SisLab:3196 %I VNU University of Engineering and Technology %T Domain-independent Intent Extraction from Online Texts %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3196/ %X Identifying user’s intents from texts on online channels has wide range of applications from entrepreneurship, banking to e-commerce. However, intent identification is not a simple task due to intent and its attributes are various and strongly depend on the domain of data. In our research, we study the problem of domain-independent intent identification from posts and comments crawled from social networks and discussion forums. We present ten general labels, i.e. labels do not depend on a specific domain, and utilize them when extracting intent and its related information. We also propose the map between general labels and domain-specific labels. We extensively conduct experiments to explore the efficiency of using general labels compared to specific labels in extracting user’s intents when the number of intent domains increases. Our study is conducted on a medium-sized dataset from three selected domains: Tourism, Real Estate and Transportation. In term of accuracy, when the number of domains grows, our proposal achieves significantly better results than the domain-specific method in identifying user’s intent. %0 Conference Paper %A Tran, Phuong Nam %A Ta, Van Duc %A Truong, Quoc Tuan %A Duong, Quang Vu %A Nguyen, Thac Thong %A Phan, Xuan Hieu %B ACIIDS: the 8th Asian Conference on Intelligent Information and Database Systems %C Da Nang, Vietnam %D 2016 %F SisLab:1639 %P 170-180 %T Named Entity Recognition for Vietnamese Spoken Texts and Its Application in Smart Mobile Voice Interaction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1639/ %V 9621 %0 Conference Paper %A Tran, Phuong Nhung %A Nguyen, Cam Tu %A Chun, Jinhee %A Le, Vu Ha %A Tokuyama, Takeshi %B 2013: International Conference on Knowledge Discovery and Information Retrieval (KDIR) %D 2013 %F SisLab:851 %T A Multiple Instance Learning Approach to Image Annotation with Saliency Map %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/851/ %0 Conference Paper %A Tran, Quan Hung %A Nguyen, Le Minh %A Pham, Bao Son %B 2014: The 6th International Conference on Knowledge and Systems Engineering (KSE 2014) %C Hanoi, Vietnam %D 2014 %F SisLab:449 %P 641-651 %T Question Analysis for a Community-Based Vietnamese Question Answering System %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/449/ %0 Conference Paper %A Tran, Quoc Long %A Farajtabar, Mehrdad %A Le, Song %A Zha, Hongyuan %B Individual Activities 2015 SIAM International Conference on Data Mining %D 2015 %F SisLab:1717 %T NetCodec: Community Detection %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1717/ %0 Conference Paper %A Tran, Quoc Long %A Farajtabar, Mehrdad %A Song, Le %A Zha, Hongyuan %B Proceedings of the 2015 SIAM International Conference on Data Mining %D 2015 %F SisLab:3268 %P 91-99 %T Netcodec: Community detection from individual activities %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3268/ %0 Journal Article %A Tran, Quoc Long %A Toh, K-A %A Srinivasan, Dipti %A Wong, K-L %A Low, Shaun Qiu-Cen %D 2005 %F SisLab:3273 %I IEEE %J IEEE Transactions on Systems, Man, and Cybernetics, Part B (Cybernetics) %N 5 %P 1079-1091 %T An empirical comparison of nine pattern classifiers %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3273/ %V 35 %0 Conference Paper %A Tran, Quoc Long %A Toh, Kar-Ann %A Srinivasan, Dipti %B IEEE Conference on Cybernetics and Intelligent Systems, 2004. %D 2004 %F SisLab:3269 %P 982-986 %T Adaptation to changes in multimodal biometric authentication %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3269/ %V 2 %0 Conference Paper %A Tran, Quoc Long %A Walker, LD %A Mazencev, R %A Jones, C %A Gray, A %A Fernandez, F %A McDonald, J %A Benigno, B %B INTERNATIONAL JOURNAL OF GYNECOLOGICAL CANCER %D 2015 %F SisLab:3277 %N 9 %P 1293-1293 %T SPECTROMETRIC METABOLIC PROFILING OF BLOOD SERA DETECTS EARLY STAGE OVARIAN CANCER WITH HIGH ACCURACY %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3277/ %V 25 %0 Conference Paper %A Tran, Thai Son %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1554 %T Nghiên cứu nâng cao hiệu quả và độ bảo mật cho các hệ thống nhúng %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1554/ %X Hệ thống nhúng (embedded system) được định nghĩa là một hệ thống chuyên dụng, có khả năng tự hành và được thiết kế tích hợp vào một hệ thống lớn hơn để thực hiện các chức năng chuyên biệt. Hệ thống nhúng bao gồm cả thiết bị phần cứng và phần mềm, thỏa mãn yêu cầu hoạt động theo thời gian thực. Hệ thống nhúng đặc trưng áp dụng trong lĩnh vực xây dựng và dân dụng là hệ thống điều khiển, giám sát và quản lý tòa nhà BMS (Building Management System). BMS bao gồm tổ hợp của các hệ thống nhúng thành phần và việc ứng dụng các hệ thống này giúp tăng tiện nghi cho người sử dụng, giảm chi phí vận hành của tòa nhà, tiết kiệm năng lượng và bảo vệ môi trường. Nghiên cứu này tập trung vào vấn đề bảo mật, vấn đề tối ưu hóa nguồn năng lượng và hiệu năng của hệ thống nhúng trong quản lý, giám sát tòa nhà %0 Conference Paper %A Tran, Thanh Huyen %A Cao, Duc Thinh %A Yamada, Koichi %A Nguyen, Do Van %B Joint 10th International Conference on Soft Computing and Intelligent Systems and 19th International Symposium on Advanced Intelligent Systems in conjunction with Intelligent Systems Workshop 2018 (SCIS&ISIS 2018) %D 2018 %F SisLab:3295 %T Incremental Updating Methods with Three-way Decision Models in Incomplete Information Systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3295/ %0 Journal Article %@ 1875-8681 %A Tran, Thanh Luong %A Ha, Quang Thuy %A Hoang, Thi Lan Giao %A Nguyen, Linh Anh %A Nguyen, Hung Son %D 2014 %F SisLab:1301 %J Fundamenta Informaticae %N 2 %P 287-303 %T Bisimulation-Based Concept Learning in Description Logics %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1301/ %V 133 %0 Conference Paper %A Tran, Thanh Luong %A Ha, Quang Thuy %A Hoang, Thi Lan Giao %A Nguyen, Linh Anh %A Nguyen, Hung Son %A Szalas, Andrzej %A VNU-UET, %B 2012 Fourth International Conference on Knowledge and Systems Engineering (KSE) %C Danang, Vietnam %D 2012 %F SisLab:108 %K Educational institutions;Electronic mail;Information systems;Knowledge based systems;Machine learning;Standards;Boolean functions;bisimulation equivalence;information systems;learning (artificial intelligence);Boolean attribute;bisimulation;concept constructor;concept learning method;concept name;data role;description logic-based information system;discrete attribute;functionality;machine learning;numeric attribute;unquantified number restriction;bisimulation;concept learning;description logic;machine learning; %P 65 -73 %T Concept Learning for Description Logic-Based Information Systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/108/ %X The work [1] by Nguyen and Szalas is a pioneering one that uses bisimulation for machine learning in the context of description logics. In this paper we generalize and extend their concept learning method [1] for description logic-based information systems. We take attributes as basic elements of the language. Each attribute may be discrete or numeric. A Boolean attribute is treated as a concept name. This approach is more general and much more suitable for practical information systems based on description logic than the one of [1]. As further extensions we allow also data roles and the concept constructors "functionality" and "unquantified number restrictions". We formulate and prove an important theorem on basic selectors. We also provide new examples to illustrate our approach. %0 Conference Paper %A Tran, Thanh Luong %A Ha, Quang Thuy %A Hoang, Thi Lan Giao %A Nguyen, Linh Anh %A Nguyen, Hung Son %A Szalas, Andrzej %A VNU-UET, %B 2012: 4th International Conference on Knowledge and Systems Engineering (KSE) %C Danang, Vietnam %D 2012 %F SisLab:83 %P 65-73 %T Concept Learning for Description Logic-based Information Systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/83/ %0 Conference Paper %A Tran, The Hung %A Nguyen, Van Hung %A Hoang, Anh Minh %B KSE: the 2015 International Conference on Knowledge and Systems Engineering %C Ho Chi Minh city, Vietnam %D 2015 %F SisLab:1715 %T Fast Approximate near Neighbor Algorithm by Clustering in High Dimensions %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1715/ %0 Journal Article %@ 1314-4081 %A Tran, Thi Oanh %A Dang, Hai Trieu %A Dinh, Viet Thuong %A Truong, Thi Minh Ngoc %A Vuong, Thi Phuong Thao %A Phan, Xuan Hieu %D 2017 %F SisLab:2545 %J Cybernetics and Information Technologies %N 2 %P 164-182 %T Performance Prediction for Students: A Multi-Strategy Approach %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2545/ %V 17 %X This paper presents a study on Predicting Student Performance (PSP) in academic systems. In order to solve the task, we have proposed and investigated different strategies. Specifically, we consider this task as a regression problem and a rating prediction problem in recommender systems. To improve the performance of the former, we proposed the use of additional features based on course-related skills. Moreover, to effectively utilize the outputs of these two strategies, we also proposed a combination of the two methods to enhance the prediction performance. We evaluated the proposed methods on a dataset which was built using the mark data of students in information technology at Vietnam National University, Hanoi (VNU). The experimental results have demonstrated that unlike the PSP in e-Learning systems, the regression-based approach should give better performance than the recommender system-based approach. The integration of the proposed features also helps to enhance the performance of the regression-based systems. Overall, the proposed hybrid method achieved the best RMSE score of 1.668. These promising results are expected to provide students early feedbacks about their (predicted) performance on their future courses, and therefore saving times of students and their tutors in determining which courses are appropriate for students’ ability. %0 Conference Paper %A Tran, Trong Hieu %A Nguyen, Thi Hong Khanh %A Ha, Quang Thuy %A Vu, Ngoc Trinh %B ACIIDS: the 8th Asian Conference on Intelligent Information and Database Systems %C Da Nang, Vietnam %D 2015 %F SisLab:1720 %T Argumentation framework for merging stratified belief bases %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1720/ %0 Conference Paper %A Tran, Trong Hoa %A Vu, Quang Dung %A Nakajima, Nobuyasu %B 2014: The 2nd International Conference on Computing, Management and Telecommunications (ComManTel 2014) %C Danang, Vietnam %D 2014 %F SisLab:1197 %P 269-274 %T A Method of Collecting and Analyzing Web Data Automatically %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1197/ %0 Journal Article %A Tran, Truc Mai %A Duong, Le Minh %D 2013 %F SisLab:779 %J Tạp chí Bưu chính Viễn thông %T Cải tiến cho điều khiển tốc độ tầng giao vận trong mạng điều khiển không dây đa bước %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/779/ %V 29 %0 Journal Article %A Tran, Truc Mai %A Duong, Le Minh %D 2013 %F SisLab:183 %J Chuyên san Bưu chính Viễn thông %N 29 %T MAC-based Improvement for Transport Rate Control in Multihop Wireless Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/183/ %0 Journal Article %A Tran, Truc Mai %A Duong, Le Minh %D 2013 %F SisLab:781 %J Tạp chí công nghệ thông tin và truyền thông %T Điều khiển tốc độ dựa trên thông số tầng MAC cho giao thức tầng giao vận trong mạng không dây đa bước %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/781/ %V 29 %0 Conference Paper %A Tran, Tuan %A Khanh, Nguyen %A Le, Son %A Phan, Anh %A Mizuhito, Ogawa %A Nguyen, Minh %B 2018 10th International Conference on Knowledge and Systems Engineering (KSE) (KSE'18) %D 2018 %F SisLab:3205 %K malware detection; malware analysis; deep learning; convolutional neural network %T Comparison of Three Deep Learning-based Approaches for IoT Malware Detection %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3205/ %X The development of IoT brings many opportunities but also many challenges. Recently, increasing more malware has been appeared to target IoT devices. Machine learning is one of the typical techniques used in the detection of malware. In this paper, we survey three approaches for IoT malware detection based on the application of convolutional neural networks on different data representations including sequences, images, and assembly code. The comparison was conducted on the task of distinguishing malware from nonmalware. We also analyze the results to assess the pros/cons of each method. %0 Conference Paper %A Tran, Tuan Vinh %A Nguyen, Thanh Thuy %A Nguyen, Thi Nhat Thanh %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1555 %T Nghiên cứu và ứng dụng ảnh LiDAR trong giám sát sol khí phát thải từ nông nghiệp %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1555/ %X Ô nhiễm không khí là một trong những vấn đề nghiêm trọng nhất trên thế giới, đặc biệt là tại các nước đang phát triển. Để kiểm soát mức độ ô nhiễm không khí, một trong những phương pháp được áp dụng là phân loại sol khí và các thành phần gây ô nhiễm, từ đó phát hiện các nguồn phát thải chính của chúng để có biện pháp xử lý. Gần đây, kỹ thuật LiDAR phát triển được dùng trong nghiên cứu các quá trình hỗn loạn, các chu trình trong khí quyển, các hiện tượng khí tượng học như bão, áp thấp, các luồng gió dưới tác động của núi. LiDAR cho phép theo dõi sự thay đổi của mật độ khí thải, các lỗ thủng tầng ozone, cung cấp dữ liệu về ảnh hưởng của các hiệu ứng thời tiết, các đám bụi núi lửa và sự lan truyền xuyên châu lục của khí thải, bụi sa mạc, và khói của các vụ cháy rừng. Định hướng của đề tài này là tập trung vào các thuật toán xử lý nguồn dữ liệu LiDAR để phân loại sol khí nói chung và sol khí có nguồn gốc từ các hoạt động nông nghiệp và áp dụng trong giám sát ô nhiễm không khí. %0 Conference Paper %A Tran, Tuan Vinh %A Pham, Van Ha %A Nguyen, Thanh Thuy %A Nguyen, Thi Nhat Thanh %B The 2020 12th International Conference on Knowledge and Systems Engineering (KSE) %D 2020 %F SisLab:4322 %P 73-78 %T Analysis of CALIPSO satellite imagery for air pollution source identification in Hanoi, Vietnam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4322/ %0 Book Section %A Tran, Tuan Vinh %A Pham, Van Ha %A Pham, Xuan Thanh %A Nguyen, Xuan Anh %A Bui, Quang Hung %A Nguyen, Thanh Thuy %A Nguyen, Thi Nhat Thanh %B Land-Atmospheric Research Applications in South and Southeast Asia %D 2017 %F SisLab:2767 %I Springer Nature %T Satellite Aerosol Optical Depth over Vietnam, an analysis from VIIRS and CALIOP aerosol products. %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2767/ %0 Report %9 Technical Report %@ TR2018-FIT-16 %A Tran, Van Lien %A Nguyen, Viet Thang %A Quach, Cong Hoang %A Phan, Xuan Hieu %A Pham, Minh Trien %D 2018 %F SisLab:3195 %I VNU University of Engineering and Technology %T Depth camera based navigation algorithms for indoor mobile robot %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3195/ %X How can we efficiently search an object in a room? This report introduces a method for a single indoor mobile robot to find a hidden item based on states of the room when the robot is moving. A 2D distribution, called cognitive map, is built during robot movements to boost the exploring time. It is known that in the filed of exploring algorithms, A∗ usually takes more time to reach the target than recent invented algorithms such as rapidly-exploring random trees (RRT) and probabilistic roadmap (PRM). However, by adapting the cognitive map as a cost map, the A∗ algorithm is significantly improved and surpasses the two algorithms in Scannet 3D dataset. We also introduce application of depth sensors and SLAM solvers on reconstructing the room and updating cognitive map. By running a virtual robot in Gazebo simulator, it is proved that our method can work well on synthetic environment and hence, is very promising to be worked on real-life environment. %0 Conference Paper %A Tran, Van Manh %A Vu, Thi Hong Nhan %B IIHMSP/FITAT %F SisLab:4286 %T Design of architecture for carpooling based on flexible pickup and delivery locations %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4286/ %0 Conference Paper %A Tran, Viet Hong %A Pham, Anh Tuan %A Nguyen, Van Vinh %A Nguyen, Hoai Xuan %A Nguyen, Huy Quang %B KSE: the 6th International Conference on Knowledge and Systems Engineering %C Hanoi, Vietnam %D 2014 %F SisLab:1413 %T Parameter Learning for Statistical Machine Translation using CMA-ES %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1413/ %0 Book %A Tran, Xuan Tu %A Phan Hai Phong, %D 2020 %E Tran, Xuan Tu %F SisLab:3956 %I Vietnam National University Publishing House %P 235 %T Mạng trên chip %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3956/ %0 Book %A Tran, Xuan Tu %A Le, Anh Cuong %A Tran, Duc Tan %A Pham, Manh Thang %A Do, Thi Huong Giang %D 2016 %F SisLab:2062 %I VNU University of Engineering and Technology %N 1 %P 56 %T The 2016 Annual Scientific Report of VNU University of Engineering and Technology %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2062/ %V 1 %X The University of Engineering and Technology (UET) was founded in 2004 as a member university of Vietnam National University, Hanoi (VNU). The last only five years have seen immense progresses in the development of our university in various aspects of its activities thanks to the great efforts put in by the government with strong financial supports, by the VNU with steering leadership and administration, by the national and international educational institutions and corporations in fruitful collaborations and advices, and by all the staff and students of the UET with determination, hard work and patience. Today, our university has built up a solid foundation for a sustainable and dynamic development in coming years. This solid foundation consists of a completed system of well-designed programs at both undergraduate and graduate levels, renovated teaching and learning facilities, a just limited but strong and advanced research infrastructure with state-of-the-art and completed equipment systems, a sustainable and fruitful collaboration with leading research and educational institutions in the country, and a bright, young, well-educated and welltrained faculty who are eager to take on advanced research and studies. All these have made our university a prestigious institution and firmly established its education and research capabilities. This is proved by the facts that only in the last four years our university has established close and solid ties and cooperation with prestigious foreign universities, just to name a few, the University Paris-Sud 11 (France), Japan Advanced Institute of Science and Technology, the National University of Singapore, Nanyang Technological University (Singapore), University of New South Wales and with famous international industrial and technological corporations such as Samsung, Toshiba, NEC, Mitani, Human Resocia, IBM, and other well-known institutions in the region and around the world. This is also proved in a convincing way by the facts that only in the last two years, our university has been granted several significant research projects from various national research programs. Our university has become an attractive destination for talented students at all training levels. We are proud that our university is the only Vietnamese higher-education institution that has one among the 100 best student’s teams all over the world participating the World Final of the ACM International Collegiate Programming Contest (rank #29 in 2015). We are also proud that in this year, our faculty has won an award in the National Contest in Information Technology entitled “Vietnamese Talent Award” and three awards in the VNU Science & Technology Award for the period of 2011-2015. In the coming period, we shall put our great efforts to solidify our strength and prestige. We shall widen our training scope while paying significant attention to further improvement on the education quality. We shall strongly enhance our concentrated research activity by exploring efficiently and effectively our well established foundation for a further dynamic development and, thus, contributing our crucial part to the socio-economic development of our nation and fulfilling our great missions clearly assigned by the government in the decision on the establishment of our university. To do that, innovative thinking and acting by all faculty, administrative staff and students are of deciding role and of crucial importance. %0 Patent %@ 1-0021424 (1-2017-00868) %A Tran, Xuan Tu %A Nguyen, Ngoc Sinh %A Bui, Duy Hieu %A SISLAB, VNU-UET, %D 2019 %F SisLab:3545 %T Quy trình mã hóa liên khung hình hỗ trợ xác định khối ảnh lặp lại, giảm kích thước chuỗi bit sau mã hóa và loại bỏ hiệu ứng do sai số lượng tử cho khối ảnh lặp lại %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3545/ %X Sáng chế đề cập đến quy trình mã hóa liên khung hình (inter-frame coding) cho phép xác định các khối ảnh lặp lại. Sau đó, thực hiện một quy trình mã hóa riêng cho các khối ảnh lặp lại nhằm giảm kích thước chuỗi bit mã hóa và loại bỏ các ảnh hưởng tiêu cực của sai số lượng tử. Bên cạnh đó, sáng chế cũng trình bày một số giải pháp để tăng tốc công đoạn xác định khối ảnh lặp lại. %0 Conference Paper %A Tran, Xuan Tu %A Phan, Hai Phong %A Tran, Van Huan %A Tran, Quang Vinh %A Nguyen, Ngoc Binh %A VNU-UET, %B The IEICE VLSI Design Technologies (VLD) Conference %C Okinawa, Japan %D 2010 %F SisLab:56 %P 169-174 %T Design and Implementation of an AMBA AHB Compliant Bus Architecture on FPGA %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/56/ %X To meet the increasing demands of recent applications, systems-on-chips (SoCs) are more and more complex and one system can be composed of many computing resources. The communication solution between these resources becomes one of big challenges in the design of SoCs. In this paper, we present the design and implementation of an AMBA Advanced High-performance Bus based bus architecture for a SoC platform on a Xilinx Virtex-4 XC4VLX40 FPGA. This architecture includes bus master and bus slave wrappers in order to ease the integration of intellectual property cores in systems. Simulation and implementation results are also reported and analyzed to prove the performance of the designed architecture. %0 Book %A Tran, Xuan Tu %A Phan, Xuan Hieu %A Tran, Duc Tan %A Pham, Manh Thang %A Pham, Duc Thang %A Pham, Bao Son %A VNU-UET, %C Hanoi, Vietnam %D 2017 %E Tran, Xuan Tu %E Pham, Bao Son %F SisLab:2561 %I VNU-UET %P 35 %T Annual Scientific Report 2017 (ASR 2017) %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2561/ %X The VNU-UET Annual Scientific Report has been started from 2016. %0 Conference Paper %A Tran, Xuan Tuyen %A Dinh, Tran Hiep %A Le, Vu Ha %A Zhu, Qiuchen %A Ha, Quang %B The 2020 IEEE/ASME International Conference on Advanced Intelligent Mechatronics (AIM 2020) %D 2020 %F SisLab:3996 %T Defect detection based on singular value decomposition and histogram thresholding %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3996/ %X This paper presents a novel method for defect detection based on singular value decomposition (SVD) and histogram thresholding. First, the input image is divided into blocks, where SVD is applied to determine if a region contains crack pixels. The detected crack blocks are then merged to construct a histogram to calculate the best binarization threshold by incoporating a recent technique for multiple peaks detection and Otsu algorithm. To validate the effectiveness and advantage of the proposed approach over related thresholding algorithms, experiments on images collected by an unmanned aerial vehicle have been conducted for surface crack detection. The obtained results have confirmed the merits of the proposed approach in terms of accuracy when using some well-known evaluation metrics. %0 Journal Article %@ 2169-3536 %A Tran Dinh, Duong %A Bui Duy, Dang %A Ogata, Kazuhiro %F SisLab:4565 %I IEEE %J IEEE Access %T Simulation-based invariant verification technique for the OTS/CafeOBJ method %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4565/ %X We demonstrate the power of the simulation-based invariant verification technique through two case studies in which it is formally verified that two mutual exclusion protocols, MCS protocol and Anderson protocol, enjoy the mutual exclusion property by the simulation-based invariant verification technique. We initially attempted formally verifying that the two protocols enjoy the property by the induction-based invariant verification technique. We successfully completed the formal proof for MCS protocol by the simulation-based invariant verification technique earlier than the one by the induction- based invariant verification technique even though we started the latter earlier than the former. We did not successfully complete the formal proof for Anderson protocol by the induction-based invariant verification technique as of the time of this paper submission. We define a variant of simulation, "observably equivalent simulations," and prove some theorems on them to conduct the MCS case study. Given a state machine M, another state machine M' constructed by introducing auxiliary variables into M observably equivalently simulates M. In this sense, M' can be regarded as an abstract version of M. %0 Conference Paper %A Tran Dinh, Duong %A Waki, Kentaro %A Ogata, Kazuhiro %B International Conference on Software Engineering and Knowledge Engineering %C Wyndham Pittsburgh University Center, Pittsburgh, USA %F SisLab:4566 %T Formal specification and model checking of a recoverable wait-free version of MCS %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4566/ %X MCS is widely known as one of the most efficient and influential spinning lock mutual exclusion protocols. The protocol, however, only works under the assumption that processes do not crash while acquiring/releasing the lock or being in the critical section. Furthermore, the exit segment pseudo-code of MCS’s algorithm is not wait-free since a process releasing the lock needs to wait for the next process in the virtual queue to perform some steps. A new version of MCS has been proposed by S. Dhoked and N. Mittal such that the new version is wait-free and recoverable (i.e., if some processes crash, the protocol can recover and work normally). In this paper, we formally specify the recoverable wait-free version of MCS and conduct model checking to check whether the protocol enjoys the mutual exclusion property. Our experiments say that: (1) the property is not satisfied if crashes are allowed to occur without any restriction, (2) the protocol enjoys the property if crashes never happen at all, or (3) if crashes have not occurred recently. We also describe the challenge of how to formally specify dynamic memory allocation and present our solution to solve that problem. %0 Conference Paper %A Tran Hong, Viet %A Nguyen Hoang, Quan %A Nguyen Van, Vinh %A 1University of Economic and Technical Industries,Hanoi, Vietnam, %A VNU University of Engineering and Technology, %A VNU University of Engineering and Technology, %B Cicling 2018 %D 2018 %F SisLab:3381 %T A Neural Network Classifier Based on Dependency Tree for English-Vietnamese Statistical Machine Translation %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3381/ %X Reordering in MT is a major challenge when translating between languages with different of sentence structures. In Phrase-based statistical machine translation (PBSMT) systems, syntactic pre-ordering is a commonly used preprocessing technique. This technique can be used to adjust the syntax of the source language to that of the target language by changing the word order of a source sentence prior to translation and solving to overcome a weakness of classical phrase-based translation systems: long distance reordering. In this paper, we propose a new pre-ordering approach by defining dependency-based features and using a neural network classifier for reordering the words in the source sentence into the same order in target sentence. Experiments on English-Vietnamese machine translation showed that our approach yielded a statistically significant improvement compared to our prior baseline phrase-based SMT system. %0 Conference Paper %A Tran Viet, Khoa %A Yuris Mulya, Saputra %A Dinh Thai, Hoang %A Nguyen Linh, Trung %A Nguyen N, Diep %A Nguyen Viet, Ha %A Eryk, Dutkiewicz %B IEEE Wireless Communications and Networking Conference (WCNC) %C Seoul, South Korea %D 2020 %F SisLab:4000 %T Collaborative learning model for cyberattack detection systems in IoT industry 4.0 %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4000/ %X Although the development of IoT Industry 4.0 has brought breakthrough achievements in many sectors, e.g., manufacturing, healthcare, and agriculture, it also raises many security issues to human beings due to a huge of emerging cybersecurity threats recently. In this paper, we propose a novel collaborative learning-based intrusion detection system which can be efficiently implemented in IoT Industry 4.0. In the system under consideration, we develop smart "filters" which can be deployed at the IoT gateways to promptly detect and prevent cyberattacks. In particular, each filter uses the collected data in its network to train its cyberattack detection model based on the deep learning algorithm. After that, the trained model will be shared with other IoT gateways to improve the accuracy in detecting intrusions in the whole system. In this way, not only the detection accuracy is improved, but our proposed system also can significantly reduce the information disclosure as well as network traffic in exchanging data among the IoT gateways. Through thorough simulations on real datasets, we show that the performance obtained by our proposed method can outperform those of the conventional machine learning methods. %0 Conference Paper %A Trieu, Hai Long %A Dang, Thanh Quyen %A Nguyen, Phuong Thai %A Nguyen, Le Minh %B IWSLT: the 2015 International Workshop on Spoken Language Translation %C Da Nang, Vietnam %D 2015 %F SisLab:1713 %T Phrase-based Machine Translation System %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1713/ %0 Journal Article %@ 0866-8612 %A Trieu, Hai Long %A Nguyen, Phuong Thai %A Nguyen, Le Minh %D 2015 %F SisLab:1714 %I VNU Journal of Science %J VNU Journal of Computer Science and Communication Engineering %N 1 %P 32-44 %T A New Feature to Improve Moore's Sentence Alignment Method %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1714/ %V 31 %0 Conference Paper %A Trinh, Thanh Binh %A Truong, Ninh Thuan %A Nguyen, Viet Ha %B KSE: the 2011 International Conference on Knowledge and Systems Engineering %C Vietnam %D 2011 %F SisLab:1768 %P 143-149 %T Refining Undetermined Events for Specifying Concurrent Programs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1768/ %X Development of many system features in a robust specification is a challenge in reactive system design. This paper presents a formal approach for specifying concurrent programs by refining undetermined events in Event-B. This approach investigates the concurrent aspect in specification and provides a general structure of models using Event-B specification. According to this structure, we illustrate the specification of three concurrent mechanisms: critical section access, reader-writer and producer-consumer. It is shown that the concurrent aspects in programming can be specified coherently using Event-B refinement. %0 Journal Article %A Trinh, Tuan Duong %A Nguyen, Ngoc Linh %D 2019 %F SisLab:3654 %I Hanoi University of Industry %J Journal of Science and Technology %T Ứng dụng bộ lọc kalman trong việc xử lý tín hiệu thu được từ cảm biến đo nồng độ bụi SHARP GP2Y1010AU0F %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3654/ %X Bài báo trình bày việc áp dụng thuật toán lọc Kalman trong việc xử lý tín hiệu thu thập được từ cảm biến đo nồng độ bụi Sharp GP2Y1010AU0F. Kết quả được kiểm chứng trên một hệ thống được thiết kế gồm: Một module có nhiệm vụ thu thập và biến đổi dữ liệu thu được từ cảm biến đo nồng độ bụi; một module có nhiệm vụ xử lý và lọc dữ liệu này nhờ bộ lọc Kalman, rồi gửi dữ liệu lên máy tính thông qua giao tiếp RS232. Hai module giao tiếp không dây với nhau bằng sóng vô tuyến 2.4 GHz. This paper presents the application of using Kalman filtering to filter the data which collected from Sharp GP2Y1010AU0F dust level sensor. A real system combined by two modules communicating with each other through a 2.4 GHz radio frequency line is designed to apply the method. One module collects and transforms data while the other one filters data with a Kalman filter then transfers this data to PC. %0 Conference Paper %A Truong, Anh Hoang %A Dang, Van Hung %A Dang, Duc Hanh %A Vu, Xuan Tung %B 12th International Conference on Distributed Computing and Internet Technology (ICDCIT) %D 2016 %F SisLab:1953 %I Springer %P 157-168 %T A Type System for Counting Logs of Multi-threaded Nested Transactional Programs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1953/ %V 9581 %0 Journal Article %A Truong, Anh Hoang %A Nguyen, Ngoc Khai %D 2015 %F SisLab:1739 %I Trường Đại học Sư phạm Hà Nội %J Tạp chí khoa học %T Hệ thống kiểu tính cận trên số log cho ngôn ngữ giao dịch đa luồng tối giản %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1739/ %0 Conference Paper %A Truong, Anh Hoang %A Nguyen, Ngoc Khai %B Hội nghị nghiên cứu và ứng dụng Công nghệ thông tin 2015 %C Hanoi, Vietnam %D 2015 %F SisLab:1740 %T Hệ thống kiểu để tính biên tài nguyên của các chương trình giao dịch lồng nhau và đa luồng %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1740/ %0 Conference Paper %A Truong, Anh Hoang %A Nguyen, Ngoc Khai %A Dang, Van Hung %A Dang, Duc Hanh %B 13th International Conference on Theoretical Aspects of Computing (ICTAC) %D 2016 %F SisLab:1954 %I Springer %P 82-99 %T Calculating Statically Maximum Log Memory Used by Multi-threaded Transactional Programs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1954/ %V 9965 %0 Journal Article %A Truong, Anh Hoang %A Tran, Huy Vu %A Nguyen, Bao Ngoc %D 2014 %F SisLab:1191 %J VNU Journal of Science: Comp. Science & Com. Eng. %N 3 %P 12-21 %T Symbolic round-off error between floating-point and fixed-point %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1191/ %V 30 %0 Conference Paper %A Truong, Dac Duy %A Nguyen, Ngoc Hoa %B FAIR'10 - Fundamental and Applied IT Research. %C Da Nang %D 2017 %F SisLab:2677 %P 449-458 %T MÔ HÌNH THU THẬP VÀ QUẢN LÝ DỮ LIỆU TRONG CÁC HỆ THỐNG GIÁM SÁT VÀ ĐIỀU KHIỂN CÔNG NGHIỆP ỨNG DỤNG ĐIỆN TOÁN ĐÁM MÂY %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2677/ %X Một trong những mục tiêu quan trọng của cách mạng công nghiệp lần thứ 4 (industrial 4.0) là đưa các thiết bị vật lý sử dụng trong công nghiệp có khả năng tự động truyền tải dữ liệu, trạng thái lên mạng Internet và tiếp nhận điều khiển mà không cần đến sự can thiệp của con người. Để làm được việc đó, một trong những bài toán cần phải giải quyết là xây dựng một phương pháp thu thập và quản lý dữ liệu phù hợp trong đó các dữ liệu trạng thái, hoạt động của thiết bị cơ khí sẽ được xử lý và lưu trữ trên các cơ sở dữ liệu đám mây (cloud database). Qua đó, người dùng sẽ có thế trực tiếp quan sát, giám sát và điều khiển các thiết bị thông qua trình duyệt Web, ứng dụng điện thoại thông minh,…, độc lập về mặt địa lý mà không cần thiết phải hiện diện trong môi trường làm việc của thiết bị. Chính vì những yêu cầu và lợi ích trên, trong bài báo này, chúng tôi giới thiệu kết quả nghiên cứu, xây dựng và đề xuất mô hình thu thập và quản lý dữ liệu của các thiết bị sử dung trong công nghiệp cho hệ thống quản lý, giám sát và điều khiển được xây dựng trên đám mây. Thử nghiệm đã được tiến hành trên hệ thống giám sát và điều khiển các trạm trộn bê tông cho phép khẳng định những lợi điểm của mô hình đề xuất so với những hệ thống SCADA truyền thống. %0 Journal Article %@ 0218-1940 %A Truong, Ninh Thuan %A Nguyen, Viet Ha %D 2013 %F SisLab:767 %J International journal of software engineering and knowledge engineering (IJSEKE) %T An Approach to Checking Security Policy in Software Development %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/767/ %0 Conference Paper %A Truong, Ninh Thuan %A Nguyen, Viet Ha %B 14th Asia-Pacific Software Engineering Conference (APSEC'07) %D 2007 %F SisLab:1774 %P 199-205 %T Using B to Verify the Weaving of Aspects %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1774/ %X Aspect J is an aspect-oriented extension of the Java language that enables a modular implementation of crosscutting concerns. Despite this, aspects lack support for formal specification and verification. This paper expresses the base class and some related aspects of Aspect J model in B notation. It aims to benefit from proof obligations generated by B tools to ensure the correctness of Aspect J component composition. Static crosscuts of aspects are guaranteed by proof obligations of relation clauses between B abstract machines and dynamic crosscuts are proved by proof obligations of B refinement machines. This approach is illustrated by verifying a simple example. %0 Conference Paper %A Truong, Ninh Thuan %A Trinh, Thanh Binh %A Nguyen, Viet Ha %B Seventh IEEE International Conference on Software Engineering and Formal Methods %C Hanoi, Vietnam %D 2009 %F SisLab:1778 %P 201-209 %T Coordinated Consensus Analysis of Multi-agent Systems Using Event-B %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1778/ %X Formal specifications and reasoning techniques in software modelling are needed to ensure the correctness of the system at the design phase. Event-B is a formal method with support tools that allows the stepwise development of reactive systems. Such systems include multi-agent systems as a subclass. In this paper, we propose an approach to specify capabilities of a number of software agents. We then verify whether these capabilities help the agents to accomplish a certain task using a supported tool for Event-B. We use the binary numeral system as a case study to illustrate our approach. %0 Book Section %A Trương, Anh Hoàng %A Nguyen, Van Vinh %B Thư viện thông minh 4.0: Công nghệ - Dữ liệu - Con người %D 2018 %F SisLab:3206 %I VNU %T Nâng cao hiệu quả sách giáo dục với công nghệ %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3206/ %X Sự phát triển mạnh và có nhiều thành tựu đột phá của của công nghệ thông tin và truyền thông trong những năm gần đây đã và đang và sẽ tiếp tục thay đổi cách con người sống, làm việc, và giải trí. Trong lĩnh vực giáo dục nói chung và đặc biệt là sách giáo khoa và sách tham khảo nói riêng, việc khai thác ứng dụng công nghệ thông tin trong giáo dục đã được chú ý từ lâu. Sách điện tử theo dạng số hóa sách giấy, có thể thêm một số thông tin đa phương tiện, đã không còn xa lại với đa số người dùng. Tuy nhiên sách điện tử dạng này chưa khai thác được nhiều tiềm năng của công nghệ có thể giúp học sinh, sinh viên khai thác sách hiệu quả hơn. Trong bài này chúng tôi phân tích sự phát triển của ứng dụng công nghệ trong sách giáo dục và ứng dụng công nghệ trong học tập để từ đó đưa ra một số đề xuất để giúp việc phát triển các sách giáo dục nói chung và đặc biệt sách giáo khoa nói riêng nhằm nâng cao giá trị cho sách, giúp người học tiếp thu kiến thức hiệu quả hơn mà tác giả cũng không mất thêm quá nhiều công sức. Các đề xuất này cũng áp dụng cho các thư viện truyền thống, có kho nội nội dung rất phong phú của sách giấy nhưng ngày càng tỏ ra hạn chế để thu hút người học đến khai thác. %0 Journal Article %@ 1572-8102 %A Tung, Vu Xuan %A To, Van Khanh %A Ogawa, Mizuhito %D 2017 %F SisLab:2543 %I Springer %J Formal Methods in System Design %P 1-38 %T raSAT: an SMT solver for polynomial constraints %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2543/ %X This paper presents raSAT SMT solver, which is aimed to handle polynomial constraints over both reals and integers with simple unified methodologies. Its three main features are (1) a raSAT loop for inequalities, which adds testing to interval constraint propagation to accelerate SAT detection, (2) a non-constructive reasoning for equations over reals based on the generalized intermediate value theorem, and (3) soundness of floating-point arithmetic that is guaranteed by (a) rounding up/down over-approximations of intervals, and (b) confirmation of a satisfying instance detected by testing using the iRRAM package, which guarantees error bounds. %0 Journal Article %@ 1568-1577 %A Van Damme, Ellen %A Laukens, Kris %A Dang, Thanh Hai %A Van Ostade, Xaveer %D 2010 %F SisLab:2074 %J International Journal of Biological Sciences %N 1 %P 51-67 %T A manually curated network of the PML nuclear body interactome reveals an important role for PML-NBs in SUMOylation dynamics %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2074/ %V 6 %0 Conference Paper %A Varró, Dániel %A Asztalos, Márk %A Bisztray, Dénes %A Boronat, Artur %A Dang, Duc Hanh %A Geiß, Rubino %A Greenyer, Joel %A Gorp, Pieter Van %A Kniemeyer, Ole %A Narayanan, Anantha %A Rencis, Edgars %A Weinell, Erhard %B The 3rd International Conference on Applications of Graph Transformations with Industrial Relevance (AGTIVE) %D 2007 %F SisLab:1966 %I Springer %P 540-565 %T Transformation of UML Models to CSP: A Case Study for Graph Transformation Tools %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1966/ %V 5088 %0 Conference Paper %A Vo, Chi My %A Nguyen, Tri Thanh %A Nguyen, Quoc Long %A Bui, Khac Luyen %B 2014: the 3nd International Conference on Advances in Mining and Tunneling (ICAMT) %C Vung Tau, Viet Nam %D 2014 %F SisLab:1310 %T Applicability of Artificial Neural Networks on Surface Subsidence Prediction due to Underground Mining %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1310/ %0 Patent %@ 1-2018-05587 %A Vo, Dinh Hieu %A Nguyễn, Ngọc Sơn %A Trần, Minh Tuấn %A Nguyễn, Văn Sơn %D 2020 %F SisLab:4353 %T Quy trình kiểm tra trùng lặp trong nhóm văn bản %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4353/ %0 Conference Paper %A Vo, Hieu Dinh %A Phung, Dung Chi %A Vu, Quang Dung %A Nguyen, Viet Ha %A VNU-UET, %B 2012 Fourth International Conference on Knowledge and Systems Engineering (KSE) %C Danang, Vietnam %D 2012 %F SisLab:107 %K Cryptography;Privacy;Protocols;Runtime;Service oriented architecture;Web services;information systems;security of data;service-oriented architecture;specification languages;BPEL;SOA;administrative domains;composite Web services;data security;information systems;security issues;service composition;service-oriented architecture;BPEL;SOA;security;service composition; %P 60 -64 %T Securing Data in Composite Web Services %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/107/ %X Service-oriented architecture (SOA) provides a solution for building information systems with reusability, flexibility, and extensibility. In SOA, new services can be developed by using existing services. BPEL is considered as the main means for composing services. In service composition, a composite service may be built from element services which belong to different administrative domains. Therefore, in order to exploit the benefits of service composition, we must carefully consider security issues. One of them is to protect users' information. In this paper, we describe an approach for securing BPEL composite Web services. We assume that some users' data received by the composite services needs securing. The proposed approach makes sure that when the data is sent out from the composite service appropriate security policies will be applied. %0 Conference Paper %A Vo, Quoc Bao %A Tran, Trong Hieu %A Nguyen, Thi Hong Khanh %B Knowledge-Based and Intelligent Information & Engineering Systems: Proceedings of the 20th International Conference KES-2016, York, UK, 5-7 September 2016. %D 2016 %F SisLab:3193 %P 295-304 %T On the Use of Surplus Division to Facilitate Efficient Negotiation in the Presence of Incomplete Information %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3193/ %0 Conference Paper %A Vo, Van Hoang %A Nguyen, Nam Hai %A Nguyen, Ngoc Hoa %B Kỷ yếu Hội nghị Quốc gia lần thứ XI về Nghiên cứu cơ bản và ứng dụng Công Nghệ thông tin (FAIR) %C Hà Nội %D 2018 %F SisLab:3165 %T MỘT SỐ KỸ THUẬT GÂY RỐI, CHE DẤU MÃ ĐỘC MACRO %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3165/ %X Tấn công mạng ngày càng diễn biến phức tạp, đa dạng về hình thức và chủng loại. Đặc biệt đối với các đối tượng người dùng khi sử dụng hệ thống thư điện tử để trao đổi, là một trong những mục tiêu mà Hacker hướng đến. Các văn bản được giao dịch ngày càng nhiều thông qua hệ thống thư điện tử, nó dường như không thể thiểu trong thời buổi công nghệ chính phủ điện tử hiện nay, đó cũng là một trong những điểm yếu dẫn đến các cuộc tấn công mạng vào người dùng, khi Hacker sử dụng các kỹ thuật che dấu mã độc vào file văn bản để tấn công máy tính người dùng thông qua hệ thống thư điện tử. Đây cũng là một trong những hình thức tấn công phổ biến hiện nay được Hacker sử dụng, kẻ tấn công không cần phải tạo ra những chủng loại mã độc mới, phức tạp, mà chỉ cần thay đổi làm rối các dấu hiệu nhận dạng, để tránh sự phát hiện của các chương trình, thiết bị an ninh. Bài báo này đề xuất phương pháp mới gây rối mã độc và kết hợp các kỹ thuật gây rối nhằm mục đích che dấu mã độc để lẩn tránh sự trước sự phát hiện của các chương trình bảo vệ như IDS/IPS, AV…được sử dụng thông qua tính năng macro trong chương trình Office của Microsoft. Đây là một trong những trình soạn thảo văn bản phổ biến nhất và được sử dụng rộng rải trong việc giao dịch qua hệ thống thư điện tử. Đó là sự kết hợp các kỹ thuật làm rối mã độc macro, khi kết hợp các kỹ thuật này, cho phép kẻ tấn công che dấu mã độc trước phần lớn sự phát hiện của các chương trình, thiết bị bảo vệ, đặc biệt là các thiết bị phát hiện và ngăn chặn mã độc như Anti-virus. Để từ đó giúp người dùng, và người quản trị mạng có cái nhìn rõ hơn, đồng thời có những giải pháp phòng tránh trước các dạng tấn công này %0 Journal Article %A Vu, Ba Duy %A Du, Phuong Hanh %D 2014 %F SisLab:1415 %J Tạp chí Đại học Quốc gia Hà Nội %N 3 %P 1-6 %T Nâng cao hiệu năng xử lý giao tác với mô hình CSDL trong bộ nhớ %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1415/ %V 30 %0 Journal Article %@ 1745-1361 %A Vu, Dieu Huong %A Chiba, Yuki %A Yatake, Kenro %A Aoki, Toshiaki %D 2015 %F SisLab:1473 %I IEICE %J IEICE Transactions on Information and Systems %N 6 %P 1137-1149 %T A Framework for Verifying the Conformance of Design to Its Formal Specifications %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1473/ %V E98-D %0 Conference Paper %A Vu, Dieu Huong %A Chiba, Yuki %A Yatake, Kenro %A Aoki, Toshiaki %A Yuki, %B TASE 2016 %C Shanghai, China %D 2016 %F SisLab:2379 %P 81-88 %T Verifying OSEK/VDX OS Design using Its Formal Specification %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2379/ %X Automotive systems are widely used in industry and our daily life. As the reliability of automotive systems is becoming a greater challenge in our community, increasingly more automotive companies are interested in applying formal methods to improve the reliability of automotive systems. We focus on automotive operating systems conforming to the OSEK/VDX standard. Such operating systems are considered as important components to ensure the reliability of the automotive systems. In previous work, we proposed a framework to verify the design models of reactive systems against their specifications. This framework allows us to check whether the design model conforms to the specification based on a simulation relation. This paper shows a case study in which the framework is applied to a real design of the OSEK/VDX operating system. As a result, we found that we were able to check several important properties of the design model. We show the effectiveness and practicality of the framework based on the results of the case study. %0 Conference Paper %A Vu, Dieu Huong %A Truong, Anh Hoang %A Chiba, Yuki %A Aoki, Toshiaki %B 2017 4th National Foundation for Science and Technology Development Conference on Information and Computer Science %C Ha Noi, Viet Nam %D 2017 %F SisLab:2711 %T Automated testing reactive systems from Event-B model %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2711/ %X We present a model-based testing approach for reactive systems where both test inputs and expected results are generated from `restricted' Event-B specifications. We show that it is possible to automatically build the restricted Event-B specifications from the original ones base on a knowledge base of the system under tests. The restricted models are to reduce the state space of the original Event-B models while preserving the possible testing paths, so that our model-based generated test suite can archive equivalent path coverage as using the original models. We also present a tool and a testing skeleton that are easy to use so that system developers can effectively test an arbitrary number of scenarios with reactive systems. %0 Conference Paper %A Vu, Duc Quang %A Hoang, Xuan Huan %A Do, Thanh Mai %B FAIR 2016 %D 2016 %F SisLab:2366 %T Một thuật toán hiệu quả dựa trên giải thuật tối ưu đàn kiến giải bài toán r|p trung tâm %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2366/ %0 Conference Paper %A Vu, Duc Quang %A Nguyen, Van Truong %A Hoang, Xuan-Huan %B RIVF 2016 %D 2016 %F SisLab:2369 %T An Improved Artificial Immune Network For Solving Construction Site Layout Optimization %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2369/ %X Nature-inspired algorithms are often used to find optimal solutions for many combinatorial problems. An immune inspired algorithm, opt-aiNet algorithm, is well known for func- tion optimization. In this paper, we develop a combination of local search with opt-aiNet, called lopt-aiNet, to solve construction site layout (CSL) problem. The effectiveness of the proposed algorithm is investigated through experiments on some datasets taken from the state-of-art and a randomly created dataset. Ex- perimental results show that the lopt-aiNet can produce optimal transportation cost with lower run time compared to the site layouts generated by metaheuristics: Genetic Algorithm (GA), Particle Swarm Optimization (PSO), Ant Colony Optimization (ACO) and aiNet. %0 Conference Paper %A Vu, Duy Loi %A Nguyen, Dinh Viet %A Ta, Huu Trung %B Hội thảo Quốc gia lần thứ IX: “Một số vấn đề chọn lọc của Công nghệ thông tin và Truyền thông” %C Da Lat, Viet Nam %D 2007 %F SisLab:2987 %P 239-248 %T Nguyên tắc và hiệu suất hoạt động của mạng di động cá nhân theo tiêu chuẩn IEEE 802.15.4 %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2987/ %X Bài báo trình bày một số kết quả nghiên cứu bước đầu về mạng di động cá nhân theo tiêu chuẩn IEEE 802.15.4, một tiêu chuẩn của mạng IEEE 802.15 với tốc độ thấp, đặc biệt phù hợp để tổ chức các mạng cảm biến (sensor networks) trong tương lai. Sau khi giới thiệu tổng quan về đặc trưng công nghệ của mạng di động cá nhân IEEE 802.15, bài báo trình bày một số nguyên tắc hoạt động cơ bản của mạng di động IEEE 802.15.4, ví dụ: tổ chức kết nối ngang hàng p2p và kết nối theo hình sao, điều khiển truy nhập MAC, đặc trưng trao đổi số liệu giữa trạm điều khiển và các trạm làm việc. Một số thông số hiệu suất hoạt động của mạng như: tỷ lệ các gói tin được phân phát đến đích (packet delivery ratio), độ trễ trung bình trên một chặng (hop delay), tỷ suất liên kết thành công trạm làm việc với trạm điều khiển, ... được tính toán trên cơ sở sử dụng bộ chương trình mô phỏng mạng NS-2 cho phép bước đầu phân tích, đánh giá hiệu quả hoạt động của mạng IEEE 802.15.4. %0 Conference Paper %A Vu, Minh Manh %A Hoang, Xuan Huan %B The 2017 International Conference on Knowledge and Systems Engineering (KSE) %C Hue, Vietnam %D 2017 %F SisLab:2763 %P to appear %T Minimizing the Spread of Misinformation on online Social Networks with Time and Budget constraint %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2763/ %X In this article, we propose a linear threshold model to the problem of minimizing the spread of misinformation for cases where partial knowledge of misinformation sources is available by using monitors,the misinformation propagation time and the budget for placing monitors are restricted, and proved it is NP-Hardness. At the same time, we also suggest two greedy algorithms to solve the problem. Experimental results show the dominant advantages of the algorithms in comparison with other commonly used algorithms. %0 Conference Paper %A Vu, Ngoc Trinh %A Ha, Quang Thuy %A Nguyen, Hung Son %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1556 %T Nghiên cứu tích hợp mô hình dữ liệu trong trung tâm dữ liệu ngành dầu khí việt nam %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1556/ %X Tích hợp dữ liệu là vấn đề nghiên cứu có nhiều ứng dụng quan trọng trong Khoa học máy tính cũng như nhiều lĩnh vực của cuộc sống. Việc nghiên cứu về tích hợp dữ liệu đã được tiến hành trong nhiều thập kỷ gần đây và đã đạt được nhiều kết quả quan trọng. Tuy nhiên, với một số lĩnh vực đặc thù, các kết quả nghiên cứu về tích hợp dữ liệu còn nhiều hạn chế. Trong nghiên cứu này chúng tôi xem xét vấn đề tích hợp mô hình dữ liệu trong trung tâm dữ liệu ngành dầu khí Việt Nam. Từ việc khảo sát, đánh giá thực trạng công tác tổ chức, lưu trữ và sử dụng dữ liệu của các đơn vị trong Tập đoàn Dầu khí Việt Nam, chúng tôi xây dựng mô hình tích hợp dữ liệu cho ngành Dầu khí. Các kết quả chính đã đạt được bao gồm: a) xác định các mô hình dữ liệu từ các phần mềm quản lý dữ liệu thăm dò khai thác phổ biến hiện nay; b) xây dựng một Ontology về thăm dò khai thác dầu khí và c) đề xuất giải pháp tích hợp các ontology chuyên ngành dầu khí có sử dụng một số thuật toán học máy. %0 Conference Paper %A Vu, Ngoc Trinh %A Nguyen, Quoc Dat %A Nguyen, Tien Dat %A Nguyen, Manh Cuong %A Vu, Van Vuong %A Ha, Quang Thuy %B 2018 10th Asian Conference on Intelligent Information and Database Systems (ACIIDS) %C Dong Hoi, Vietnam %D 2018 %F SisLab:2942 %I Springer Nature 2018 %P 190-199 %T A Positive-Unlabeled Learning Model for Extending a Vietnamese Petroleum Dictionary Based on Vietnamese Wikipedia Data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2942/ %V 10751 %X This study provides a positive-unlabeled learning model for extending a Vietnamese petroleum dictionary based on Vietnamese Wikipedia data. Machine learning algorithms with positive and unlabeled data together with separated and combined between Google similarity distance and Cosine similarity distance, used in this study. The data sources used to integrate are English - Vietnamese oil and gas dictionary and the Vietnamese Wikipedia. In the results, a novelty way for data integration with higher accuracy by using a combination of algorithms. The first Vietnamese oil and gas ontology was built in Vietnam. This ontology is a useful tool for staff in the oil and gas industry in training, research, search daily. %0 Conference Paper %A Vu, Ngoc Trinh %A Tran, Van Hien %A Le, Hoang Quynh %A Tran, Mai Vu %B KSE: the 2015 International Conference on Knowledge and Systems Engineering %C Ho Chi Minh city, Vietnam %D 2015 %F SisLab:1611 %P 141-149 %T A Method for Building a Labeled Named Entity Recognition Corpus Using Ontologies %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1611/ %X Building a labeled corpus which contains sufficient data and good coverage along with solving the problems of cost, effort and time is a popular research topic in natural language processing. The problem of constructing automatic or semi-automatic training data has become a matter of the research community. For this reason, we consider the problem of building a corpus in phenotype entity recognition problem, class-specific feature detectors from unlabeled data based on over 10260 unique terms (more than 15000 synonyms) describing human phenotypic features in the Human Phenotype Ontology (HPO) and about 9000 unique terms (about 24000 synonyms) of mouse abnormal phenotype descriptions in the Mammalian Phenotype Ontology. This corpus evaluated on three corpora: Khordad corpus, Phenominer 2012 and Phenominer 2013 corpora with Maximum Entropy and Beam Search method. The performance is good for three corpora, with F-scores of 31.71% and 35.77% for Phenominer 2012 corpus and Phenominer 2013 corpus; 78.36% for Khordad corpus. %0 Journal Article %@ 2525-2348 %A Vu, Nguyen Thuc %A Le, Tuan Linh %D 2018 %F SisLab:3395 %I Tạp chí khoa học quản lý và công nghệ - Trường đại học Công Nghệ và Quản Lý Hữu Nghị %J Tạp chí khoa học quản lý và công nghệ - Trường đại học Công Nghệ và Quản Lý Hữu Nghị %T Thiết kế và xây dựng phần mềm nền tảng (Platform) cung cấp, quảng bá dịch vụ Y tế thông minh hướng tới cá nhân Structural design and the development, setting up the Platform for individualized commom uses to provide and advertise the Smart Medical Service %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3395/ %V 06 %X Trong những năm vừa qua, cùng với sự xuất hiện của các thuật ngữ “Cách mạng Công nghiệp lần thứ 4” (hay còn gọi là CMCN 4.0), Internet vạn vật (Internet of Things - IoT) và phong trào khởi nghiệp sáng tạo khắp nơi trên thế giới dựa vào Công nghệ Thông tin - Truyền thông (ICT) chúng ta đã được thấy sự xuất hiện của rất nhiều phần mềm và nền tảng làm thay đổi hoàn toàn cuộc sống của cộng đồng. Các ứng dụng này đem đến rất nhiều sự tiện lợi trong cuộc sống, làm giảm các chi phí không cần thiết, tiết kiệm thời gian, công sức của con người một cách đáng kể. Ở Việt Nam, chúng ta dễ dàng kiểm đếm một số tên tuổi nổi bật trong các lĩnh vực chính như: Grab, Go-Viet, EMDDI, FastGo, Be, Aber, XELO,… (lĩnh vực giao thông); MoMo, 1Pay, VTCPay, ViettelPay, VNPTPay, ZaloPay, SamsungPay, Vimo, Tima, … (lĩnh vực công nghệ tài chính); Now, Giaohangtietkiem.vn, LaLamove, GHN, ShipVN,… (lĩnh vực giao vận); Booking.com, Trivago, Tripi, Traveloka,… (lĩnh vực du lịch). Trong lĩnh vực Y tế cũng đã có một số ứng dụng được xây dựng và triển khai ở Việt Nam như eDoctor, Vicare, Jio Health, VieVie, Bacsi24, Udr,… Tuy nhiên, theo tìm hiểu của chúng tôi thì ở lĩnh vực Y tế chưa có ứng dụng (phần mềm) nào cung cấp nền tảng dùng chung (Platform) để các cơ sở Y tế chỉ việc dùng như là một hệ thống riêng của mình mà không phải bỏ công sức và tiền bạc ra xây dựng cũng như phải có đội ngũ cán bộ Công nghệ Thông tin chuyên trách (Đây là một vấn đề rất nan giải với các cơ sở Y tế ở quy mô còn nhỏ, đội ngũ cán bộ mỏng và năng lực quản trị vẫn còn hạn chế do ở trong giai đoạn đầu phát triển). Trong bài này, chúng tôi đề cập đến việc thiết kế cấu trúc và xây dựng một nền tảng Y tế thông minh dùng chung (Ứng dụng ViDoctor) hướng tới cá nhân %0 Conference Paper %A Vu, Quang Dung %A Nguyen, Ha Thanh %A Vu, Danh Viet %A Nguyen, Viet Ha %A Nakajima, Nobuyasu %B 2014 : The 5th Symposium on Information and Communication Technology (SoICT 2014) %C Ha Noi, Viet Nam %D 2014 %F SisLab:1198 %T An Architecture for Web Services Mash-Up based on Mobile Agents %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1198/ %0 Conference Paper %A Vu, Quang Dung %A Nguyen, Viet Ha %A Nobuyasu, Nakajima %B 2014: The 2nd International Conference on Computing, Management and Telecommunications (ComManTel 2014) %C Danang, Vietnam %D 2014 %F SisLab:1195 %P 257 -262 %T Agent based Adaptive Connection of Mobile Devices %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1195/ %0 Conference Paper %A Vu, Quang Dung %A Vu, Danh Viet %A Nguyen, Ha Thanh %A Bui, Phi Diep %A Nguyen, Viet Ha %A Nakajima, Nobuyasu %B 2014: The 2nd International Conference on Computing, Management and Telecommunications (ComManTel 2014) %C Danang, Vietnam %D 2014 %F SisLab:1196 %P 263-268 %T Adaptive Web Page Layout for Mobile Devices %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1196/ %0 Conference Paper %A Vu, Quang Hiep %A Li, Meijing %A Vu, Thi Hong Nhan %A Ryu, Keun Ho %B Conference on Awareness Science and Technology and Ubi-Media Computing %D 2013 %F SisLab:2121 %T A novel clustering method for animal trajectory analysis using Wireless Sensor Network %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2121/ %0 Conference Paper %A Vu, Quang Hiep %A Vu, Thi Hong Nhan %A Kim, Yonggil %B 6th International Conference FITAT %D 2013 %F SisLab:2123 %T Predicting the intended wildlife movement in the Wireless Sensor Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2123/ %0 Conference Paper %A Vu, Quang Hiep %A Vu, Thi Hong Nhan %A Ryu, Keun Ho %B 2014: the 7th International Conference on Frontiers of Information Technology, Applications and Tools (FITAT’14), The Fourth PT-ERC International Symposium on Personalized Medicine (FITAT & ISPM) %C Chiang Mai, Thailand %D 2014 %F SisLab:1426 %T Constructing a System for Monitoring, Managing Groundwater in the Industrial Zones of Hanoi City %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1426/ %0 Conference Paper %A Vu, Quang Hiep %A Vu, Thi Hong Nhan %A Ryu, Keun Ho %B FITAT/ISPM 2014 %C Thailand %D 2014 %F SisLab:2120 %T Constructing a System for Monitoring, Managing Groundwater in the Industrial Zones of Hanoi City %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2120/ %0 Conference Paper %A Vu, T. Thai %A Nguyen, Dinh Viet %A Nguyen, M. Hien %B 2013: International Conference on Advanced Engineering - Theory and Application (AETA 2013) %C Saigon, Vietnam %D 2014 %F SisLab:1313 %P 297-306 %T An Energy-Aware Routing Protocol for Wireless Sensor Networks Based on K-Means Clustering %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1313/ %0 Conference Paper %A Vu, Thi Dao %A Pham, Ngoc Hung %A Nguyen, Viet Ha %B SW4PHD: the 2016 Scientific Workshop for PhD Students %C Hanoi %D 2016 %F SisLab:1557 %T A Method for Automated Test Cases Generation from Sequence Diagrams and Object Constraint Language for Concurrent Programs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1557/ %X Model- based testing plays a significant role in practice and a lot of researches on it has been investigated in recent years due to great benefits. There are some approaches for model- based testing: test data generation, test cases generation from behavior models and test scripts generation from abstract tests. Current research focuses on generating executable test cases from Unified Modeling Language (UML) sequence diagrams and object constraint language (OCL). In this paper we propose an automated test cases generation method from sequence diagrams, class diagrams, and object constraint language. The method supports UML 2.0 sequence diagrams including eight kinds of combined fragments describing control follow of systems. Test cases are generated with respect to the given concurrency coverage criteria. With strong concurrency coverage, generating exhaustive test cases for all concurrent interleaving sequences is exponential in size. The key idea of this method is to create selection of possible test scenarios in special case of exploring the message sequence with their possible interleaving in parallel or weak sequencing fragments. Test data for testing loop fragments is also generated. We implemented a tool to automate the proposed method and studied its feasibility and effectiveness. Experimental results show that the method can generate test cases on demand to satisfy a given concurrency coverage criterion and can detect up to 74.5% of seeded faults. %0 Journal Article %@ 0866-8612 %A Vu, Thi Dao %A Pham, Ngoc Hung %A Nguyen, Viet Ha %D 2016 %F SisLab:1982 %I VNU Journal of Science %J VNU Journal of Computer Science and Communication Engineering %T A Method for Automated Test Cases Generation from Sequence Diagrams and Object Constraint Language for Concurrent Programs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1982/ %X This paper proposes an automated test cases generation method from sequence diagrams, class diagrams, and object constraint language. The method supports UML 2.0 sequence diagrams including eight kinds of combined fragments describing control flow of systems. Test cases are generated with respect to the given concurrency coverage criteria. With strong concurrency coverage, generating exhaustive test cases for all concurrent interleaving sequences is exponential in size. The key idea of this method is to create selection of possible test scenarios in special case of exploring the message sequence with their possible interleaving in parallel or weak sequencing fragments. Test data for testing loop fragments are also generated. We implement a tool to automate the proposed method and studied its feasibility and effectiveness. Experimental results show that the method can generate test cases on demand to satisfy a given concurrency coverage criterion and can detect up to 74. 5% of seeded faults. %0 Conference Paper %A Vu, Thi Dao %A Pham, Ngoc Hung %A Nguyen, Viet Ha %B The 9th Asian Conference on Intelligent Information and Database Systems %C Japan %D 2017 %F SisLab:2697 %P 525-536 %T A Method for Automated Test Cases Generation from UML Models with String Constraints %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2697/ %X This paper proposes an automated test cases generation method from sequence diagrams and class diagrams with string constraints. The method supports UML 2.0 sequence diagrams including twelve combined fragments. An algorithm for generating test scenarios are developed to avoid test paths explosion without having data sharing points of threads in parallel fragments or weak sequencing fragments. Test data are also generated with solving constraints of string variables. We standardize string constraints and equations at the boundary of variables that are input formula of Z3-str solver. Comparing with the current approach of the solver, some preprocessing rules are extended for other operations such as charAt, lastindexOf, trim, startsWith and endsWith. If a result of the Z3-str is SAT, test data of each test scenario are generated to satisfy the constraints with boundary coverage. A tool is implemented to support the proposed method, and some experiments are also presented to illustrate the effectiveness of the tool. %0 Conference Paper %A Vu, Thi Dao %A Pham, Ngoc Hung %A Nguyen, Viet Ha %B SoICT: the 2015 Symposium on Information and Communication Technology %C Hue, Vietnam %D 2015 %F SisLab:1540 %P 335-341 %T A Method for Automated Test Data Generation from Sequence Diagrams and Object Constraint Language %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1540/ %X This paper proposes an automated test data generation method from the information embedded in model elements such as Unified Modeling Language (UML) sequence diagrams, class diagrams, and Object Constraint Language (OCL). The method supports UML 2.0 sequence diagrams including eight kinds of combined fragments describing control flow of systems. Comparing with some approaches by using depth first search (DFS) or breadth first search (BFS) algorithms, the proposed method generates all possible test scenarios with the higher error uncover capability. Test data for testing loop fragment is also generated. Therefore, it helps to detect errors in testing loops and the concurrency errors such as safety and liveness property of the systems. %0 Journal Article %A Vu, Thi Dao %A To, Van Khanh %A Nguyen, Viet Ha %D 2014 %F SisLab:1772 %J Tạp chí CNTT & TT: Các công trình nghiên cứu, phát triển và ứng dụng CNTT & TT %N 11 %P 70-82 %T Phương pháp sinh các ca kiểm thử tự động từ các mô hình thiết kế UML và ngôn ngữ ràng buộc đối tượng OCL %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1772/ %V V %0 Conference Paper %A Vu, Thi Hong Nhan %B 6th International Conference FITAT %D 2013 %F SisLab:2122 %T An Android-based Intelligent Tourism Recommendation System %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2122/ %0 Conference Paper %A Vu, Thi Hong Nhan %B 4th International Conference FITAT/DBMI %C Korea %D 2011 %F SisLab:2132 %T A Framework for Detecting Health Condition of a Mobile Subject %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2132/ %0 Conference Paper %A Vu, Thi Hong Nhan %B Advances in Intelligent Information Hiding and Multimedia Signal Processing, Smart Innovation, Systems and Technologies %D 2019 %F SisLab:3749 %T A Framework for Ridesharing Recommendation Services %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3749/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Bui, The Duy %A Vu, Quang Hiep %A Ta, Viet Cuong %B International Conference on Convergence Technology %C Vietnam %D 2012 %F SisLab:2127 %T A Breadth-First Search Based Algorithm for Mining Frequent Movements From Spatiotemporal Databases %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2127/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Chi, Cheong Hee %A Ryu, Keun Ho %B Asian Symposium on Geographic Information Systems %C Japan %D 2005 %F SisLab:2145 %T Mining Sequential Patterns from Moving Object Databases %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2145/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Chi, Jeong Hee %A Ryu, Keun Ho %B Asian Symposium on Geographic Information System %C China %D 2004 %F SisLab:2146 %T A Data Model for Region with Indeterminate Boundaries %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2146/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Chi, Jeong Hee %A Ryu, Keun Ho %B Asia Pacific Web Conference, LNCS %C China %D 2006 %F SisLab:2142 %T Discovery of Spatiotemporal Patterns in Mobile Environment %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2142/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Chi, Jeong Hee %A Ryu, Keun Ho %B Korea Spatial Information System Society %D 2005 %F SisLab:2144 %T An Efficient Algorithm for Mining Frequent Sequences in Spatiotemporal Data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2144/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Chi, Jeong Hee %A Ryu, Keun Ho %B OpenGIS %D 2004 %F SisLab:2147 %T Query Operations for Fuzzy Spatiotemporal Databases %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2147/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Kim, Sang Ho %A Ryu, Keun Ho %B Asian Conference on Remote Sensing and International Symposium on Remote Sensing %C Korea %D 2003 %F SisLab:2149 %T A Fuzzy Spatio-Temporal Data Model and Dynamic Query Operations %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2149/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Kim, Sang Ho %A Ryu, Keun Ho %B Asia Pacific Web Conference %C China %D 2005 %F SisLab:2143 %T Representation and Manipulation of Geospatial Objects with Indeterminate Extents %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2143/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Kim, Sang Ho %A Ryu, Keun Ho %B First Asian Symposium on Geographic Information Systems from Computer Science and Engineering View %D 2003 %F SisLab:2148 %T Spatio-Temporal Data Modeling for Fuzzy Regions %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2148/ %0 Journal Article %A Vu, Thi Hong Nhan %A Lee, Bum Ju %A Ryu, Keun Ho %D 2008 %F SisLab:2136 %J Korea Information Processing Society (KIPS) journal %T Discovery of Frequent Sequence Patterns in Moving Object Databases %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2136/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Lee, Jun Wook %A Kim, Hiseok %A Ryu, Keun Ho %B IEEE 8th International Conference on Computer and Information Technology Workhops %C Australia %D 2008 %F SisLab:2135 %T An HRV Patterns Discovering Neural Network for Mobile Healthcare Services %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2135/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Lee, Yang Koo %B the ACM 5th International Conference FITAT %C Mongolia %D 2012 %F SisLab:2125 %T Development of a Mobile Tourism System %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2125/ %0 Journal Article %@ 0010-485X %A Vu, Thi Hong Nhan %A Lee, Yang Koo %A Bui, The Duy %D 2014 %F SisLab:1424 %J Computing Journal %N 3 %P 163-188 %T A Technique for Extracting Behavioral Sequence Patterns From GPS Data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1424/ %V 96 %0 Journal Article %A Vu, Thi Hong Nhan %A Lee, Yang Koo %A Bui, The Duy %D 2013 %F SisLab:785 %I Springer %J Computing Springer %N 3 %P 163-188 %T A Technique for Extracting Behavioral Sequence Patterns from GPS Recorded Data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/785/ %V 96 %0 Conference Paper %A Vu, Thi Hong Nhan %A Lee, Yang Koo %A Bui, The Duy %B the ACM 5th International Conference FITAT %C Mongolia %D 2012 %F SisLab:2124 %T Trajectory Classification Using Spatiotemporal Sequence %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2124/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Lee, Yang Koo %A Le, Thanh Ha %B 2014: International Conference on Indoor Positioning and Indoor Navigation (IPIN 2014) %C Busan, Korea %D 2014 %F SisLab:1425 %T Forecasting Moving Object Position based on Temporal Patterns %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1425/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Lee, Yang Koo %A Le, Thanh Ha %A Bui, The Duy %B ACM 4th International Conference FITAT/DBMI 2011 %C Korea %D 2011 %F SisLab:2131 %T Location-based Information Disseminating System with Integration of Various Georeferenced Data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2131/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Lee, Yang Koo %A Namsrai, Oyun-Erdene %B International Conference on Frontiers of Information Technology, Applications and Tools %D 2018 %F SisLab:2977 %T An Approach to Cab-sharing Service Development %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2977/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Lee, Yang Koo %A Oyun-Erdene, Namsrai %B The 9th International Conference on Frontiers of Information Technology, Applications and Tools %C Zhuhai, China %D 2016 %F SisLab:2065 %T Activity Recognition based on Clustering Methods for Senior Homecare Services %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2065/ %X In modern society, most seniors prolong their independence. To guarantee the safety of them when living on their own, we need to monitor their activities all the time and react to critical situations. The rapid advances in wireless networks, wearable sensors, and communications technologies pave the way for the advent of homecare service systems. Activity recognition is a crucial task in building such systems. This paper investigates two clustering methods, kmeans and Self-organizing map (SOM) for recognizing human daily activities. An experiment is performed on a real data set. The results show that k-means performs pretty well in classifying two activities; however the accuracy is pretty low when the data set is scaled up to five activities. SOM outperforms k-means in most cases of data sets. On average, the resulting accuracy of SOM is 87% and of k-means is 54% for five activities. As a result, SOM is most suitable to be integrated in systems for providing remote homecare services. %0 Conference Paper %A Vu, Thi Hong Nhan %A Lee, Yang Koo %A Vu, Quang Hiep %A Bui, The Duy %B 6th International Conference on Computer Sciences and Convergence Information Technology, IEEE %C Korea %D 2011 %F SisLab:2129 %T Mining Constraint-Based Patterns in Moving Object Database %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2129/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Lee, Yang Koo %A Vu, Quang Hiep %A Bui, The Duy %B ICCM %C Korea %D 2012 %F SisLab:2126 %T A user context recognition for ubiquitous computing systems %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2126/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Namsrai, Oyun-Erdene %A Lee, Yang Koo %B The 10th International Conference on the Frontiers of Information Technology, Applications and Tools %D 2017 %F SisLab:2691 %T A design of ridesharing recommendation system %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2691/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Nguyen, Chi Thuc %A Lee, Yang Koo %A Namsrai, Oyun-Erdene %B International conference on information information, system, and convergence applications %D 2017 %F SisLab:2693 %T Discovery of frequent routes for ridesharing services %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2693/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Park, Namkyu %B IEEE 40th International Conference on Computers & Industrial Engineering, Soft Computing Techniques for Advanced Manufacturing and Service Systems %C Japan %D 2010 %F SisLab:2134 %T Heart Rate Variability Pattern Recognition in Ambulatory Environments %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2134/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Ryu, Keun Ho %B 6th Asia Pacific International Symposium on Information Technology %C Malaysia %D 2007 %F SisLab:2137 %T An Approach to Mining Frequent Movement Patterns in Traffic Data %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2137/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Ryu, Keun Ho %B Asian Symposium on Geographic Information System %C Korea %D 2006 %F SisLab:2138 %T An Approach to Mining Mobility Association Rules %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2138/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Ryu, Keun Ho %B Korea Spatial Information System Society %C Korea %D 2006 %F SisLab:2139 %T Estimating Object Location in Location-based Services %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2139/ %0 Conference Paper %A Vu, Thi Hong Nhan %A Ryu, Keun Ho %B International Conference on Intelligent Computing, LNCS %C China %D 2006 %F SisLab:2141 %T Future Location prediction of Moving Objects Based on Movement Rules %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2141/ %0 Journal Article %A Vu, Thi Hong Nhan %A Ryu, Keun Ho %A Park, Namkyu %D 2009 %F SisLab:2117 %J Journal of Computer & Industrial Engineering %T A Method for Predicting Location of Mobile User for Location-based Services System %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2117/ %0 Journal Article %@ 0866-8612 %A Vu, Thi Ngoc Anh %A Nguyen, Trong Dong %A Nguyen, Vu Hoang Vuong %A Dang, Thanh Hai %A Do, Duc Dong %D 2019 %F SisLab:3508 %I VNU Journal of Science %J VNU Journal of Computer Science and Communication Engineering %T Adaptive Large Neighborhood Search Enhances Global Protein-Protein Network Alignment %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3508/ %0 Conference Paper %A Vu, Thi Ngoc Anh %A Nguyen, Vu Hoang Vuong %A Dang, Thanh Hai %A Do, Duc Dong %B 2017 4th NAFOSTED Conference on Information and Computer Science (NICS 2017) %C Hanoi, Vietnam %D 2017 %F SisLab:2706 %P 13-18 %T APP-RB: An efficient method for biological network querying based on rebuild procedure %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2706/ %X Biological molecules (e.g. DNAs, RNs, proteins, etc.) do not function in isolation themselves but in a sophisticated manner of various interactions between them to carry out every biological processes in living cells. Data about these interactions are exponentially increased thanks to advanced technologies. Analyzing biological interaction networks is essential to get insights into how biological system works. Among analytics tasks querying a small network of interest against a well studied large-scale one is regarded as one of the most essential task and is proven to be a NP-complete problem. This paper presents an efficient biological network querying method called APP-RB, which is an improvement of the state-of-the-art biological network queyring method APPAGATO introduced in 2016 by taking advantages of FASTAn, an PPI alignment algorithm. Experimental results showed that our algorithm performed much better than the original APPAGATO. With these results, we hope that our proposed method APP-RB will be helpful to biologists and those who are interested in network comparison problems. %0 Conference Paper %A Vu, Trong Hoa %A Neubig, Graham %A Sakti, Sakriani %A Toda, Tomoki %A Nakamura, Satoshi %B 2014: the 14th Conference of the European Chapter of the Association for Computational Linguistics %C Gothenburg, Sweden %D 2014 %F SisLab:1309 %P 128-132 %T Acquiring a Dictionary of Emotion-Provoking Event %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1309/ %0 Conference Paper %A Vu, Trung Kien %A Tran, Quoc Long %B 2018 10th International Conference on Knowledge and Systems Engineering (KSE 2018) %C Ho Chi Minh City, Vietnam %D 2018 %F SisLab:3265 %P 164-168 %T Robust Loss Functions: Defense Mechanisms for Deep Architectures %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3265/ %X Current deep learning methods and technologies have reached the level of deployment in softwares and hardwares for real-life applications. However, recent studies have shown deep learning architectures are highly vulnerable to attacks and exploitation via input perturbations. In this works, we investigate the effects of these attacks on the outputs of each layer of deep architectures and on theirs performance in term of classification accuracy measures. The results show that without defense mechanism, even simple attacks devastated deep architectures’ outputs in every layer and theirs classification performance. We then propose multiple defense mechanisms in order to protect deep architectures and to make them more robust to input perturbation attacks. %0 Report %9 Technical Report %A Vu, Van Phong %A Nguyen, Huy Tuyen %A Nguyen, Manh Hung %A Nguyen, Ha Thanh %D 2018 %F SisLab:3194 %T A virtual player for Vietnamese version of "Who Want to Be a Millionaire" %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3194/ %X Human are trying to create intelligent systems, the creation of systems that can win intellectual games are problems researched for a long time. With the goal of building a system that is capable of answering the questions of "Who want to be a millionaire", we find that the main challenge of the system is how to build the knowledge source as well as the use of its data to serve for making the most accurate decision in the game. In this article, we focus on proposing an architecture, a method to build a part of the system having question answering function of the "Who want to be a millionaire" program by exploring the source of knowledge from Wikipedia, besides, we install and test the method based on that architecture and evaluate the results obtained. %0 Conference Paper %A Vu, Van Phuc %A Pham, Van Huong %A Nguyen, Ngoc-Binh %B 2014: Hội nghị quốc gia lần thứ VII về Nghiên cứu cơ bản và ứng dụng Công nghệ thông tin (FAIR 2014) %C Thừa Thiên Huế, Việt Nam %D 2014 %F SisLab:1194 %T Tối ưu điện năng tiêu thụ của hệ thống nhúng dựa trên kỹ nghệ ngược và tái cấu hình CPU %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1194/ %0 Conference Paper %A Vu, Xuan Son %A Le, Duc Trong %A Edlund, Christoffer %A Jiang, Lili %A Nguyen, Hoang D. %B The 28th ACM International Conference on Multimedia %D 2020 %F SisLab:4260 %T Privacy-Preserving Visual Content Tagging using Graph Transformer Networks %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4260/ %X With the rapid growth of Internet media, content tagging has become an important topic with many multimedia understanding applications, including efficient organisation and search. Nevertheless, existing visual tagging approaches are susceptible to inherent privacy risks in which private information may be exposed unintentionally. The use of anonymisation and privacy-protection methods is desirable, but with the expense of task performance.Therefore, this paper proposes an end-to-end framework (SGTN) using Graph Transformer and Convolutional Networks to significantly improve classification and privacy preservation of visual data. Especially, we employ several mechanisms such as differential privacy based graph construction and noise-induced graph transformation to protect the privacy of knowledge graphs. Our approach unveils new state-of-the-art on MS-COCO dataset in various semisupervised settings. In addition, we showcase a real experiment in the education domain to address the automation of sensitive document tagging. Experimental results show that our approach achieves an excellent balance of model accuracy and privacy preservation on both public and private datasets. Codes are available at https://github.com/ReML-AI/sgtn. %0 Conference Paper %9 Working Paper %A Vu, Xuan Son %A Nguyen, Thanh Son %A Le, Duc Trong %A Jiang, Lili %B Proceedings of the 28th International Conference on Computational Linguistics %D 2020 %F SisLab:3967 %P 414-425 %T Multimodal Review Generation with Privacy and Fairness Awareness %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3967/ %X Users express their opinions towards entities (e.g., restaurants) via online reviews which can bein diverse forms such as text, ratings, and images. Modeling reviews are advantageous for userbehavior understanding which, in turn, supports various user-oriented tasks such as recommen-dation, sentiment analysis, and review generation. In this paper, we propose MG-PriFair, a multi-modal neural-based framework, which generates personalized reviews with privacy and fairnessawareness. Motivated by the fact that reviews might contain personal information and sentimentbias, we propose a novel differentially private (dp)-embedding model for training privacy guar-anteed embeddings and an evaluation approach for sentiment fairness in the food-review domain.Experiments on our novel review dataset show that MG-PriFair is capable of generating plausiblylong reviews while controlling the amount of exploited user data and using the least sentiment-biased word embeddings. To the best of our knowledge, we are the first to bring user privacy andsentiment fairness into the review generation task. The dataset and source codes are available athttps://github.com/ReML-AI/MG-PriFair. %0 Conference Paper %A Vu, Xuan Son %A Vu, Thanh %A Nguyen, Huong %A Ha, Quang Thuy %B The 6th International Conference on Knowledge and Systems Engineering (KSE 2014) %C Hanoi, Vietnam %D 2014 %F SisLab:442 %P 233-245 %T Improving Text-Based Image Search with Textual and Visual Features Combination %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/442/ %0 Conference Paper %A Vu, Xuan Tung %A Hoang, Truong %B PAS: the 4th International Seminar on Program Verification, Automated Debugging and Symbolic Computation %C Beijing, China %D 2015 %F SisLab:1741 %T An Improved Type System for Counting Logs of Transactional Multi-threaded Programs %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1741/ %X This paper presents an improved type system to estimate an upper bound for the resource consumption of programs with nested and multi-threaded transactions. This work improves our previous type system to which in comparison the new one is more precise. In addition, the soundness of the system is sketched by the help of rewriting techniques. %0 Conference Paper %A Vu, Xuan Tung %A To, Van Khanh %A Ogawa, Mizuhito %B The 8th International Joint Conference on Automated Reasoning %D 2016 %F SisLab:2153 %I Springer-Verlag New York, Inc. %P 228-237 %T raSAT: An SMT Solver for Polynomial Constraints %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2153/ %0 Journal Article %@ 0866-8612 %A Vu Thi Ngoc, Anh %A Thai Dinh, Phuc %A Nguyen, Hoang Duc %A Dang, Thanh Hai %A Do Duc, Dong %D 2017 %F SisLab:2892 %I VNU Journal of Science %J VNU Journal of Computer Science and Communication Engineering %T Ant colony optimization based founder sequence reconstruction %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2892/ %X Reconstruction of founder (ancestor) genes for a given population is an important problem in evolutionary biology. It involves finding a set of genes that can combine together to form genes of all individuals in that population. Such reconstruction can be modeled as a combinatorial optimization problem, in which we have to find a set of founder (gene) sequences so that the individuals in a given population can be generated by the smallest number of recombination on these founder sequences. In this paper we propose a novel ant colony optimization algorithm (ACO) based method, equipped with some important improvements, for the founder gene sequence reconstruction problem. The proposed method yields excellent performance when validating on 108 test sets from three benchmark datasets. Comparing with the best by far method for founder sequence reconstruction, our proposed method performs better in 45 test sets, equally well in 44 and worse only in 19 sets. These experimental results demonstrate the efficacy and perspective of our proposed method. %0 Report %9 Technical Report %A Vuong, Thi Hai Yen %D 2019 %F SisLab:3739 %T Formal Specification and Model Checking of Some Variants of Anderson Mutual Exclusion Protocol %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3739/ %0 Conference Paper %A Vuong, Thi Hai Yen %A Nguyen, Thi Thu Trang %A Tran, Nhu Thuat %A Nguyen, Le Minh %A Phan, Xuan Hieu %B The 11th International Conference on Knowledge and Systems Engineering (KSE 2019) %C Da Nang, Vietnam %D 2019 %F SisLab:3663 %T Learning to Transform Vietnamese Natural Language Queries into SQL Commands %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3663/ %X In the field of data management, users traditionally manipulates their data using structured query language (SQL). However, this method requires an understanding of relational database, data schema, and SQL syntax as well as the way it works. Database manipulation using natural language, therefore, is much more convenient since any normal user can interact with their data without a background of database and SQL. This is, however, really tough because transforming natural language commands into SQL queries is a challenging task in natural language processing and understanding. In this paper, we propose a novel two–phase approach to automatically analyzing and converting natural language queries into the corresponding SQL forms. In our approach, the first phase is component segmenta- tion which identifies primary clauses in SQL such as SELECT, FROM, WHERE, ORDER BY, etc. The second phase is slot– filling that helps extract sub–components for each primary clause such as SELECT column(s), SELECT aggregation operation, etc. We carefully conducted an empirical evaluation for our method using conditional random fields (CRFs) on a medium– sized corpus of natural language queries in Vietnamese, and have achieved promising results with an average accuracy of more than 90%. %0 Conference Paper %A Vuong, Thi Hai Yen %A Nguyen, Van Hop %A Duong, Quang Vu %A Phi, Thi Thu %A Nguyen, Thac Thong %A Truong, Quoc Tuan %A Phan, Xuan Hieu %B SSC: the 2016 Student Scientific Contest %C Hanoi %D 2016 %F SisLab:1494 %T On the design and development of a mobile virtual assistant for Vietnamese (VAV) %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1494/ %0 Conference Paper %A Vuong, Thi Hong %A Nguyen, Minh Duc %A Nguyen, Thi Cam Van %A Tran, Mai Vu %B SSC: the 2016 Student Scientific Contest %C Hanoi %D 2016 %F SisLab:1500 %T Hệ thống tư vấn xã hội dựa trên phân tích ý định và hành vi người sử dụng %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1500/ %0 Journal Article %A Vuong, Thi Hong %A Nguyen, Thi Cam Van %A Ha, Quang Thuy %F SisLab:4290 %J Asian Conference on Intelligent Information and Database Systems %T N-tier machine learning-based architecture for DDoS attack detection %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4290/ %V 13 %X Distributed Denial of Service (DDoS) attack is a menace to network security that aims at exhausting the target networks with malicious traffic. With simple but powerful attack mechanisms, it introduces an immense threat to the current Internet community. In this paper, we propose a novel multi-tier architecture intrusion detection model based on a machine learning method that possibly detects DDoS attacks. We evaluate our model using the newly released dataset CICDDoS2019, which contains a comprehensive variety of DDoS attacks and address the gaps of the existing current datasets. Experimental results indicated that the proposed method is more efficient than other existing ones. The experiments demonstrated that the proposed model accurately recognize DDoS attacks outperforming the state-of-the-art by F1-score. %0 Conference Paper %A Vuong, Thi Hong %A Pham, Thanh Huyen %A Nguyen, Tri Thanh %A Ha, Quang Thuy %B 11th Asian Conference on Intelligent Information and Database Systems (ACIIDS 2019) %C Yogyakarta, Indonesia %D 2019 %F SisLab:2961 %I Springer %T An Adversarial Learning and Canonical Correlation Analysis based Cross-Modal Retrieval Model %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2961/ %X An Adversarial Learning and Canonical Correlation Analysis based Cross-Modal Retrieval Model %0 Conference Paper %A Vuong, Thi Hong %A Tran, Van Hien %A Nguyen, Minh Duc %A Nguyen, Thi Cam Van %A Pham, Thanh Huyen %A Tran, Mai Vu %B Knowledge and Systems Engineering (KSE) %D 2016 %F SisLab:2521 %P 264-267 %T Social-spam profile detection based on content classification and user behavior %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2521/ %X Web-based social system enables new community-based opportunities for participants to engage, share and interact. The rapid growth of Facebook has triggered a dramatic increase in spam volume and sophistication. Spammers post their status or comment in Page to send spam content to their friends or other users in the network. In this paper, we consider the problem of detecting spam accounts on Facebook based on comment content and user social behavior. We will propose a hybrid approach using Maximum Entropy (Maxent) model for classifying user comments as either spam or non-spam. We carefully conducted an empirical evaluation for our model on a large collection of comments in Vietnamese Facebook Pages and achieved promising results with an average accuracy of more than 90%. %0 Journal Article %A Vuong, Thi Phuong Thao %A Phan, Xuan Hieu %D 2017 %F SisLab:2546 %J VNU Journal of Science: Education Research %N 1 %P 10-22 %T Tiếp cận đa chiều về quản lý hoạt động đào tạo ở Đại học Quốc gia Hà Nội %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2546/ %V 33 %X Theo tiếp cận đa chiều về đảm bảo chất lượng giáo dục, bài báo này khảo sát, đánh giá công tác quản lý đào tạo đại học ở Đại học Quốc gia Hà Nội một cách toàn diện (đa tham số) và khách quan (đa phương pháp, đa nguồn). Trong đó, đánh giá của cán bộ được thực hiện qua 8 nhân tố với 48 tiêu chí bao gồm: mục tiêu và kế hoạch đào tạo; quản lý chương trình đào tạo; tuyển sinh; tổ chức thực hiện đào tạo; quản lý hoạt động giảng dạy của giảng viên; quản lý hoạt động học tập của sinh viên; quản lý đội ngũ cán bộ quản lý, nhân viên hỗ trợ hoạt động đào tạo; quản lý môi trường học tập, cơ sở vật chất và trang thiết bị. Đánh giá của sinh viên được thể hiện qua 4 nhân tố: nội dung chương trình đào tạo; hoạt động giảng dạy; hoạt động đào tạo; môi trường, cơ sở vật chất, thiết bị phục vụ giảng dạy và học tập. Các kết quả nghiên cứu đã được củng cố, khẳng định qua so sánh với kết quả đánh giá kiểm định điều kiện đảm bảo chất lượng của các chuyên gia mạng lưới các đại học ASEAN. Tám khuyến nghị được đưa ra để tăng cường điều kiện đảm bảo chất lượng đào tạo. %0 Journal Article %@ 1751-9659 %A Vuong, Tung Long %A Le, Dinh Minh %A Hoang, Van Xiem %A Dinh, Trieu Duong %A Vu, Huu Tien %A Le, Thanh Ha %D 2018 %F SisLab:3289 %I The Institution of Engineering and Technology %J IET Image Processing %T View Synthesis Method for 3D Video Coding Based on Temporal and Inter View %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3289/ %0 Conference Paper %A Vuong, Tung Long %A Le, Dinh Minh %A Le, Thi Thuy %A Le, Thanh Ha %B The 2016 International Conference on Electronics, Information and Communication (ICEIC) %C Danang, Vietnam %D 2016 %F SisLab:2034 %T Pre-rendered subtitles removal in video sequences using text detection and inpainting %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2034/ %0 Conference Paper %A Vuong, Tung Long %A Nguyen, Do Van %A Nguyen, Tai Long %A Bui, Cong Minh %A Kieu, Hai Dang %A Ta, Viet Cuong %A Tran, Quoc Long %A Le, Thanh Ha %B Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence %D 2019 %F SisLab:3742 %T Sharing Experience in Multitask Reinforcement Learning %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3742/ %0 Conference Paper %A Vương, Thị Hồng %A Nguyen, Thi Cam Van %A Ha, Quang Thuy %A VNU-UET, %D 2021 %F SisLab:4484 %I Springer %P 375-385 %T N-Tier Machine Learning-Based Architecture for DDoS Attack Detection %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4484/ %X Distributed Denial of Service (DDoS) attack is a menace to network security that aims at exhausting the target networks with malicious traffic. With simple but powerful attack mechanisms, it introduces an immense threat to the current Internet community. In this paper, we propose a novel multi-tier architecture intrusion detection model based on a machine learning method that possibly detects DDoS attacks. We evaluate our model using the newly released dataset CICDDoS2019, which contains a comprehensive variety of DDoS attacks and address the gaps of the existing current datasets. Experimental results indicated that the proposed method is more efficient than other existing ones. The experiments demonstrated that the proposed model accurately recognize DDoS attacks outperforming the state-of-the-art by F1-score. %0 Conference Paper %A Xiao, Han %A Reda, Elbarougy %A Masato, Akagi %A Junfeng, Li %A Ngo, Thi Duyen %B The 2015 RISP International Workshop on Nonlinear Circuits, Communications and Signal Processing %D 2015 %F SisLab:1602 %P 86 - 89 %T A study on perception of emotional states in multiple languages on Valence-Activation approach %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1602/ %X Human beings can judge emotional states of a voice only by listening, no matter thay understand the language or not. Investigating the commonalities and differences of emotional states perception among multiple languages is important to understand how human beings perceive emotional states among multiple languages, and to build a human perceptionmodel independent to used languages. This paper investigates commonalities and differences among multiple languages in human perception of emotional states based on a dimensional approach. The results suggest that human beings can perceive emotional states regardless of languages. Moreover, the results can be used to build a human perceptionmodel regardless of languages only by control the deviations between neutral voices and other emotional states. %0 Journal Article %@ doi: 10.1021/pr500409q %A Zang, Xiaoling %A Christina M., Jones %A Tran, Quoc Long %A Maria, Eugenia Monge %A Zhou, Manshui %A Walker L., DeEtte %A Roman, Mezencev %A Alexander, Gray %A John F., McDonald %A Facundo M., Fernández %D 2014 %F SisLab:1183 %J Journal of Proteome Research %N 7 %P 3444-3457 %T Feasibility of Detecting Prostate Cancer by Ultraperformance Liquid Chromatography-Mass Spectrometry Serum Metabolomics %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/1183/ %V 13 %0 Journal Article %A Zang, Xiaoling %A Jones, Christina M %A Tran, Quoc Long %A Monge, Maria Eugenia %A Zhou, Manshui %A Walker, L DeEtte %A Mezencev, Roman %A Gray, Alexander %A McDonald, John F %A Fernandez, Facundo M %D 2014 %F SisLab:3271 %I ACS Publications %J Journal of proteome research %N 7 %P 3444-3454 %T Feasibility of detecting prostate cancer by ultraperformance liquid chromatography--mass spectrometry serum metabolomics %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3271/ %V 13 %0 Conference Paper %A Zhang, Zhiwei %A Phan, Xuan Hieu %A Horiguchi, Susumu %B The 2008 IEEE Symposium on Mining and Web, The 22nd International Conference on Advanced Information Networking and Applications (IEEE AINA) %C Okinawa, Japan %D 2008 %F SisLab:2300 %P 1223-1228 %T An Efficient Feature Selection Using Hidden Topic in Text Categorization %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/2300/ %0 Journal Article %@ 2169-3536 %A Zhu, Qiuchen %A Dinh, Tran Hiep %A Phung, Manh Duong %A Quang, Ha %D 2021 %F SisLab:4437 %I IEEE %J IEEE Access %T Hierarchical Convolutional Neural Network with Feature Preservation and Autotuned Thresholding for Crack Detection %U https://eprints.uet.vnu.edu.vn/eprints/id/eprint/4437/ %X Drone imagery is increasingly used in automated inspection for infrastructure surface defects, especially in hazardous or unreachable environments. In machine vision, the key to crack detection rests with robust and accurate algorithms for image processing. To this end, this paper proposes a deep learning approach using hierarchical convolutional neural networks with feature preservation (HCNNFP) and an intercontrast iterative thresholding algorithm for image binarization. First, a set of branch networks is proposed, wherein the output of previous convolutional blocks is half-sizedly concatenated to the current ones to reduce the obscuration in the down-sampling stage taking into account the overall information loss. Next, to extract the feature map generated from the enhanced HCNN, a binary contrast-based autotuned thresholding (CBAT) approach is developed at the post-processing step, where patterns of interest are clustered within the probability map of the identified features. The proposed technique is then applied to identify surface cracks on the surface of roads, bridges or pavements. An extensive comparison with existing techniques is conducted on various datasets and subject to a number of evaluation criteria including the average F-measure (AFβ) introduced here for dynamic quantification of the performance. Experiments on crack images, including those captured by unmanned aerial vehicles inspecting a monorail bridge. The proposed technique outperforms the existing methods on various tested datasets especially for GAPs dataset with an increase of about 1.4% in terms of AFβ while the mean percentage error drops by 2.2%. Such performance demonstrates the merits of the proposed HCNNFP architecture for surface defect inspection.