一种基于相对区分表的属性约简算法

属性约简是知识获取中的核心问题之一。为了能较高效率地获得属性约简，在Rough Set理论基础上构造出了相对区分表，将基于相对区分表的属性约简的判定算法(JRA)作为子算法并结合归纳属性约简算法的优点，设计出了基于相对区分表的归纳属性约筒算法(RA)。算例说明该算法具有较高的属性约简效率，并能取得较好的约简结果。

Attribute reduction algorithm based on relative different table

Attribute reduction is one of the key problems for the knowledge acquisition.Based on the rough set theory,therelative different table is constructed to effectively and efficiently achieve the better attribute reduction.Then the relative differenttable is combined with the heuristic knowledge to design the algorithm for judgment of attribute reduction based on the relative dif-ferent table(JRA).And then,the inductive attribute reduction algorithm based on the relative different table is designed withthe JRA algorithm(RA).RA is illustrated by the examples.The result shows that RA can efficiently achieve the better attributereduction.