基于Rough集理论的模糊值属性信息表简化方法

为了有效地在信息表中处理取值为模糊术语的属性，解决Rough集对模糊值属性处理能力较弱的问题，提出了模糊不可分辨关系的概念，用于处理属性值为模糊术语的信息表．将约简、核、相对约简与相对核以及规则的约简与核等Rou曲集理论中一系列知识约简的概念推广到模糊环境下，提出了一种有效的模糊值信息表简化的启发式算法．数值实验验证该方法在模糊值属性信息表简化方面比传统的Pawlak方法和其他一些学者的相关工作更为有效．

Simplification of Information Table with Fuzzy-Valued Attributes Based on Rough Sets

In order to effectively deal with imprecise linguistic terms in information tables and to improve the ability of dealing with initial fuzzy data, fuzzy indiscernibility relation is proposed. Some basic concepts of rough sets such as reduct and core of knowledge are generalized to the fuzzy environment. A heuristic algorithm is presented to simplify decision tables with fuzzy-valued attributes. A numerical example is given to prove that the proposed approach is suprior to the Pawlak's typical approach and some other scholars' works in the simplification of decision tables with fuzzy-valued attributes.