基于粗糙集理论的值约简及规则提取

Data reduction and rule generation based on Rough set

结合粗糙集理论和分类规则支持度的概念,提出以值约简后实例的支持度尽可能大作为约简的目标,并给出一种值约简的算法.通过对实例分析表明,该算法能取得较好的效果.文中还讨论了规则集的性质,改进值约简算法得到一种基于粗糙集的规则挖掘算法.实验结果表明,该算法生成规则能够得到令人满意的分类正确率.

In this dissertation rough set theory is combined with the notion of support measurement, a new object of value reduction is proposed, that is getting the rules which have maximal support measurement. Based on the forementioned ideal an algorithm about value reduction is proposed, experiment show it is practical. We also discuss the property of rule set in this paper. And then we improve value reduction method and based on Rough Theory a rule generation algorithm is presented. Experimental results show that this algorithm can effectively classify unknown data.