eprintid: 2316
rev_number: 12
eprint_status: archive
userid: 286
dir: disk0/00/00/23/16
datestamp: 2017-01-13 02:33:47
lastmod: 2017-12-04 09:32:06
status_changed: 2017-01-13 02:33:47
type: conference_item
metadata_visibility: show
creators_name: Pham, Thi Ngan
creators_name: Nguyen, Van Quang
creators_name: Dinh, Duc Trong
creators_name: Nguyen, Tri Thanh
creators_name: Ha, Quang Thuy
creators_id: nganpt.di12@vnu.edu.vn
creators_id: ntthanh@vnu.edu.vn
creators_id: thuyhq@vnu.edu.vn
title: MASS: a semi-supervised multi-label classification algorithm with specific features
ispublished: pub
subjects: IT
divisions: fac_fit
abstract: 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.
date: 2017-04
date_type: published
publisher: Springer
full_text_status: none
pres_type: paper
publication: The 9th Asian Conference on Intelligent Information and Database Systems (ACIIDS) 2017
event_title: The 9th Asian Conference on Intelligent Information and Database Systems
event_location: Kanazawa, Japan
event_dates: 3-5 April 2017
event_type: conference
refereed: TRUE
citation:   Pham, Thi Ngan and Nguyen, Van Quang and Dinh, Duc Trong and Nguyen, Tri Thanh and Ha, Quang Thuy  (2017) MASS: a semi-supervised multi-label classification algorithm with specific features.  In: The 9th Asian Conference on Intelligent Information and Database Systems, 3-5 April 2017, Kanazawa, Japan.