基于极大相容块的不完备模糊目标信息系统的近似约简  

The Approximate Reduction in Incomplete and Fuzzy Objective Information Systems Based on Maximal Tolerance Classes

作　　者：胡宝清[1] 温彪[1] 

机构地区：[1]武汉大学数学与统计学院,湖北武汉430072

出　　处：《江西师范大学学报（自然科学版）》2015年第1期15-19,共5页Journal of Jiangxi Normal University(Natural Science Edition)

基　　金：国家自然科学基金(61179038)资助项目

摘　　要：在不完备模糊目标信息系统中引入了极大相容块的概念,定义了上下近似,并通过精确度的计算得到上下近似的定义具有合理性,同时提出了近似一致集的概念,给出了相关的粗糙集模型,以及该模型的近似约简概念及辨识矩阵的近似约简方法.The concept of maximal compatible classes is introduced into incomplete and fuzzy objective information systems. The upper and lower approximation is redefined,and the concept of approximation consistent set is proposed. It can be known that the definition is reasonable by calculating it. Meanwhile,the relevant rough set models are given. The identify of approximate reduction and a method for approximate reduction of identification matrix are got.

关 键 词：不完备模糊目标信息系统 极大相容块 辨识矩阵 上近似一致集 下近似一致集 近似约简 

分 类 号：O211.67[理学—概率论与数理统计]