粒矩阵与Rough关系矩阵  

Bit Granular Matrix and Rough Relation Matrix

作　　者：陈泽华[1] 谢刚[1] 谢克明[1] 

机构地区：[1]太原理工大学信息工程学院,山西太原030024

出　　处：《太原理工大学学报》2008年第4期379-382,共4页Journal of Taiyuan University of Technology

基　　金：山西省自然科学基金项目(2007021018)

摘　　要：针对Rough集本身的代数定义和其他各种扩展模型没有提供简捷的计算边界元素数目的算法,笔者定义了粒矩阵和Rough关系矩阵,采用二进制粒矩阵运算代替传统的等价类计算获得上、下近似,灵活高效。从集合论的角度完善了Rough集理论,并在此基础上,提出了基于粒矩阵的知识发现算法,同时举例证明了Rough包含与Rough相等的隶属度函数定义的充要条件仅仅是必要条件。Upper approximation and lower approximation in Rough set theory （RST） makes vague boundary computable, however, none of existing definition of RST brings efficient way to compute boundary. In this paper, Bit granular matrix and Rough relation matrix are proposed to complete the computation for upper and lower approximation, which are the most essential concepts in RST and the foundations for knowledge reduction and further research. The proposed algorithm is easier than the traditional algebraic one, and the definition of Rough relation matrix completes RST from the view of set theory. Furthermore, it is proved that the sufficient and necessary conditions for rough inclusion and rough equivalent are just necessary conditions.

关 键 词：粒计算 ROUGH集 粒矩阵 Rough关系矩阵 

分 类 号：TP301.6[自动化与计算机技术—计算机系统结构]