An Binary Discernibility Matrix Attribute Reduction Algorithm on Attribute Importance Heuristic Message

摘要

A binary discernibility matrix attribute reduction algorithm of incomplete decision table is introduced in this paper, which takes the importance of attribute as the heuristic message. To obtain better attribute reduction, this paper constructs a metric formula of attribute importance and gives a binary discernibility matrix to present incomplete decision table. Based on the formula and matrix an attribute reduction algorithm of incomplete decision table is introduced. This algorithm solves the problem of the attribute selection when the frequencies of attributes are equal. The result shows that this method is simple and effective.