摘要
论文主要对RoughSet理论中的属性约简问题进行了研究。从Skowron可辨识矩阵出发,通过对属性重要性及属性间依赖性的讨论,利用条件熵构造了一种一致数据属性约简的启发式算法;通过实例和UCI数据库证明了该算法的有效性;并对文献犤4犦中的错误进行了修正。
This paper mainly discusses the attribute reduction in Rough Set theory.Firstly,the paper researches the Skowron discernibility matrix,the significance and relevance of attributes.Meanwhile,we use the conditional entropy to propose a heuristic algorithm for attribute reduction;Then,the paper corrects the mistake in the reference.Finally,the experiment shows that it can get better effect and it also shows the reduction results of UCI database using this algorithm.
出处
《计算机工程与应用》
CSCD
北大核心
2004年第13期79-80,130,共3页
Computer Engineering and Applications
关键词
ROUGH
SET理论
属性约简
可辨识矩阵
一致数据
条件熵
启发式算法
Rough Set theory,attribute reduction,discernibility matrix,consistent data,conditional entropy,heuristic algorithm
