Fast Algorithm for Attribute Reduction Based on Rough Set Theory Using Binary Discernibility Matrix

Rough set is a valid mathematical tool that deals with uncertain, vague and incomplete information of decision systems. Attribute reduction is one of the issues in rough set theory, and core attributes are indispensable in the process of attribute reduction. Recently, some researchers proposed the method of binary discernibility matrix to compute the results of attribute reduction, which can not only save space, but also exploit the benefit of bit operations to improve the classification performance. In this paper, a novel definition of binary discernibility matrix is introduced, and then a new algorithm of computing the matrix is provided, whose time complexity is max{ (| || || ( ) |), pos O C U IND C  2 2 (| | | ( ) | )} pos O U IND C  . Finally, a fast algorithm for computing the result of attribute reduction based on rough set theory is proposed, and an example is used to illustrate the effectiveness of the proposed algorithm.