摘要
从粒度计算的角度对粗糙集理论的属性约简进行研究,分别基于代数方法和信息论方法定义了粒度差和粒度熵的概念,并在此基础上提出了两种新的属性约简算法。实验分析表明,这两种可靠有效的粒度计算方法都能得到信息表的最小约简,为进一步研究知识的粒度计算提供了可行的方法。
The attribute reduction of rough set is discussed from the viewpoint of granular computing,and the difference of granularity and granularity entropy are defined respectively based on the theory of algebra and information theory. Two new granular computing algorithms are presented. Simulation results show that the valid algorithms can find the minimal reduction, and feasible methods are put forward for further research on knowledge granularity computing.
出处
《计算机应用与软件》
CSCD
北大核心
2008年第4期239-240,共2页
Computer Applications and Software
关键词
粒度计算
粒度差
粒度熵
粗糙集
属性约简
Granular computing Difference of granularity Granularity entropy Rough set Attribute reduction
