Advertisement
Published in:
2017 | OriginalPaper | Chapter
Attribute Importance Degrees Corresponding to Several Kinds of Attribute Reduction in the Setting of the Classical Rough Sets
Author : Masahiro Inuiguchi
Published in: Fuzzy Sets, Rough Sets, Multisets and Clustering
Publisher: Springer International Publishing
Activate our intelligent search to find suitable subject content or patents.
Select sections of text to find matching patents with Artificial Intelligence. powered by
Select sections of text to find additional relevant content using AI-assisted search. powered by
Abstract
In this paper, we propose several attribute reduction concepts which are ordered linearly. For each attribute reduction, we give a discernibility matrix which enables to enumerate all reduced attribute sets. We define measures to evaluate the specificity of decision class and the retention ability of specificity corresponding to the proposed concepts of attribute reduction. Using those measures, attribute importance degrees are defined based on cooperative game theory. We show that the attribute importance degree is very different by the requirement to what extent we preserve the class information of objects. Finally, we describe the possible application of the attribute reduction to the group decision making and give modifications in case when decision classes are linearly ordered.