Research on Attribute Reduction Criteria in Decision-Theoretic Rough Set

Hengrong Ju,Xibei Yang,Hualong Yu,Yong Qi,Jingyu Yang

DOI: https://doi.org/10.3969/j.issn.1001-4616.2015.01.007

2015-01-01

Abstract:Attribute reduction is one of the important research issues in rough set theory. In classical Pawlak rough set, the lower and upper approximations are monotonic with respect to the set inclusion of attributes. However,in decision-theoretic rough set model,the lower and upper approximations may increase or decrease with respect to the increasing of attributes. To address this issue,from the viewpoint of optimization,fitness functions of the decision-monotonicity criteri-on,generality criterion and cost criterion have been proposed respectively. Genetic algorithm is also applied to compute reducts. The experimental results show that:the reducts based on decision-monotonicity criterion can generate more posi-tive rules;the reducts based on generality criterion can generate most positive rules;the reducts based on cost criterion can obtain lowest decision costs.