PARAMETRIC VERSION SPACE APPROACH TO CONCEPT-LEARNING

Authors
Citation
J. Neves et E. Smirnov, PARAMETRIC VERSION SPACE APPROACH TO CONCEPT-LEARNING, Journal of computer & systems sciences international, 37(5), 1998, pp. 708-728
Citations number
25
Categorie Soggetti
Computer Science Theory & Methods","Computer Science Artificial Intelligence","Computer Science Cybernetics","Computer Science Theory & Methods","Computer Science Artificial Intelligence","Computer Science Cybernetics
ISSN journal
10642307
Volume
37
Issue
5
Year of publication
1998
Pages
708 - 728
Database
ISI
SICI code
1064-2307(1998)37:5<708:PVSATC>2.0.ZU;2-P
Abstract
This paper presents a parametric version space approach to concept lea rning. The approach is intended for concept learning tasks that an cha racterized with different extents of expressiveness of the concept lan guages with respect to the example ones. This allows one to overcome t he isolated use of the basic version space approaches and to enable it to be applied in the broad spectrum of concept learning tasks.