基于属性布尔差别矩阵的求核算法  

Algorithm for Computing Core Based on Attribute Boolean Discerniblity Matrix

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,合肥230039 [3]滁州学院计算机系,安徽滁州239012

出　　处：《小型微型计算机系统》2012年第2期275-279,共5页Journal of Chinese Computer Systems

基　　金：安徽省自然科学基金项目(090412054)资助;安徽高校省级自然科学研究项目(KJ2011Z276)资助;安徽省高等学校省级优秀青年人才基金项目(2010SQRL138;2011SQRL123)资助;滁州学院科学研究项目(2010kj014B)资助

摘　　要：求核是粗糙集理论的主要研究内容之一.针对现有基于差别矩阵求核算法的不足,首先提出决策表简化的方法;然后,给出属性布尔差别矩阵的定义和基于属性布尔差别矩阵的核属性定义,并证明了该定义与基于正区域求核定义是等价的,同时设计相应求核算法;为了进一步提高求核效率和适应大数据集的处理,给出两个改进的求核算法.通过实例分析和实验比较,表明了本文所提出的求核算法是正确的高效的.The computation of core is one of the most important problems of rough set theory. To overcome the shortcomings of computing core attributes based on discerniblity matrix, firstly, the methods of simplifying decision table is proposed. And then, the deft- nition of attribute boolean discemiblity matrix and the corresponding definition of computing core are present. It is proved that the def- inition of core attribute is equitant with the one based on positive region. In order to improve the efficiency of computing core algo- rithrns and adapt to deal with huge data sets, two improved algorithms are put forward. Finally, both of the example analysis and experiment results show that the computing core algorithrns proposed are correct and efficient.

关 键 词：粗糙集 正区域 布尔差别矩阵 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]