A Global Search Algorithm for Attributes Reduction

摘要

Attributes reduction is a crucial problem in rough set application to data mining. In this paper, we introduce the Universal RED problem model, or UniRED, which transforms the discrete attributes reduction problems on Boolean space into continuous global optimization problems on real space. Based on this transformation, we develop a coordinate descent algorithm RED2.1 for attributes reduction problems. In order to examine the efficiency of our algorithms, we conduct the comparison between our algorithm RED2.1 and other reduction algorithms on some problems from UCI repository. The experimental results indicate the efficiency of our algorithm.