可信度差别矩阵及其属性约简  

Discernibility Matrix Based on Credibility and Attribute Reduction Method

作　　者：葛浩[1] 李龙澍[2] 杨传健[3] 

机构地区：[1]滁州学院电子信息工程系,安徽滁州239012 [2]安徽大学计算机学院,安徽合肥230039 [3]滁州学院计算机系,安徽滁州239012

出　　处：《四川大学学报（工程科学版）》2011年第5期146-152,共7页Journal of Sichuan University (Engineering Science Edition)

基　　金：安徽省自然科学基金资助项目(090412054);安徽高校省级自然科学研究项目(KJ2011Z276);安徽省高等学校省级优秀青年人才基金项目(2011SQRL123);滁州学院科学研究项目(2010kj014B2011kj003Z)

摘　　要：针对决策表中因存在不一致数据而造成求核和属性约简的错误,引入对象可信度的定义,并将其添加到决策表中,以区分不一致和一致对象;然后,给出差别矩阵的定义,以及基于该差别矩阵的核属性和属性约简定义,同时证明了由该定义获得的核和属性约简与正区域的核和属性约简是等价的;最后,设计属性约简算法及其改进算法,并通过实例和实验验证了算法的有效性。In order to resolve the error of computing core and attribute reduction,which was caused by the inconsistency of decision table,a definition of object credibility was introduced and added into the decision table to tell inconsistent objects and consistent objects.Then a definition of discernibility matrix and definitions of core and attribute reduction based on the discernibility matrix were put forward.It was proved that the core and attribute reduction acquired from the definitions are the core and attribute reduction based on positive region.Finally,the general algorithm and its improved algorithm for computing attribute reduction were designed,and both of example and experiment results showed that the algorithms are effective.

关 键 词：粗糙集 差别矩阵 属性约简 核属性 

分 类 号：TP181[自动化与计算机技术—控制理论与控制工程]