Change mechanism of a decision table’s decision performance caused by attribute reductions

Attribute reduction is one of the important topics in the research on rough set theory, it plays an important role in machine learning and data mining. However, is the decision performance of a decision table changed after an attribute reduction? In this paper, we analyze the change of the decision performance through using the positive-region reduction and the Shannon entropy reduction. The change principles of three decision measures caused by attribute reductions are proposed. Experimental analysis is performed for illuminating the change mechanism of a decision tablepsilas decision performance.