eprintid: 3154
rev_number: 10
eprint_status: archive
userid: 307
dir: disk0/00/00/31/54
datestamp: 2018-11-14 04:31:59
lastmod: 2018-11-14 04:31:59
status_changed: 2018-11-14 04:31:59
type: article
metadata_visibility: show
creators_name: Divroodi, Ali Rezaei
creators_name: Ha, Quang Thuy
creators_name: Nguyen, Linh Anh
creators_name: Nguyen, Hung Son
creators_id: thuyhq@vnu.edu.vn
corp_creators: Education DepartmentNoshahrIran
corp_creators: Faculty of Information TechnologyVNU University of Engineering and TechnologyHanoiVietnam
corp_creators: Division of Knowledge and System Engineering for ICTTon Duc Thang UniversityHo Chi Minh CityVietnam
corp_creators: Institute of InformaticsUniversity of WarsawWarsawPoland
title: On the possibility of correct concept learning in description logics
ispublished: pub
subjects: IT
divisions: fac_fit
abstract: 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.
date: 2018-02
date_type: published
publisher: Vietnam Journal of Computer Science
official_url: https://link.springer.com/article/10.1007%2Fs40595-017-0094-4
full_text_status: public
publication: Vietnam Journal of Computer Science
volume: 5
number: 1
pagerange: 3-14
refereed: TRUE
issn: 2196-8888
related_url_url: https://link.springer.com/article/10.1007%2Fs40595-017-0094-4
citation:   Divroodi, Ali Rezaei and Ha, Quang Thuy and Nguyen, Linh Anh and Nguyen, Hung Son  (2018) On the possibility of correct concept learning in description logics.  Vietnam Journal of Computer Science, 5  (1).   pp. 3-14.  ISSN 2196-8888     
document_url: https://eprints.uet.vnu.edu.vn/eprints/id/eprint/3154/1/18_06_On%20the%20possibility%20of%20correct%20concept%20learning%20in%20description%20logics_10.1007%252Fs40595-017-0094-4.pdf