An intensive study on rule acquisition in formal decision contexts based on minimal closed label concept lattices. Issue 3 (3rd July 2017)

• Record Type: Journal Article
• Authors: Li, Jinhai; Huang, Chenchen; Mei, Changlin; Yin, Yunqiang
• Is Part Of: Intelligent automation & soft computing. Volume 23:Issue 3(2017)
• Page Start: 519
• Page End: 533
• Publication Date: 2017-07-03
• Subjects: Concept lattice -- rough set theory -- lattice construction -- formal decision context -- rule acquisition; Artificial intelligence -- Periodicals; Intelligent control systems -- Periodicals; 003.5
• DOI: 10.1080/10798587.2016.1212509
• Languages: English
• ISSNs: 1079-8587

Abstract

Abstract: Rule acquisition is one of the main purposes in the analysis of formal decision contexts. Up to now, there have been several types of rules in formal decision contexts. However, these existing rules were investigated independently and are, to some extent, unsatisfactory for making better decision analysis since some of them (e.g. decision implications) are computationally expensive and others (e.g. decision rules and granular rules) are short of compactness. Motivated by these problems, minimal closed label concept lattice is defined to present limitary decision implications, which not only are easier to be extracted than decision implications, but also have more concise premises than decision rules and granular rules. Moreover, we discuss the pairwise inclusion and inference relationships among limitary decision implications, decision implications, decision rules and granular rules. Finally, some numerical experiments are conducted to demonstrate that the proposed limitary decision implications are easier to be extracted than decision implications.