COMBINING ROUGH SET WITH CERTAINTY RATIO BASED ALGORITHM

Abstract

Rough set theory (RS) is one of the important methods that concerned with analysis of data tables for the classification purposes. In this paper, a certainty ratio based algorithm is proposed. The proposed algorithm uses certainty ratio as criterion for selecting the attribute values that have a good certainty ratio, and ignore the attribute values that has low certainty level. This algorithm is combined with rough set theory in order to enhance the generated rules.  The experiments show that the proposed approach can effectively increase the accuracy of the generated rules.