关键词: 模糊粗糙集, 模糊相似度量, 模糊等价关系, 信息熵

Abstract: Fuzzy rough set，which combines the ideas of fuzzy set and rough set，is a new soft computing tool dealing with fuzzy and uncertain information.In this paper the authors propose an entropy-based knowledge acquisition approach to handle information system whose attribute values are fuzzy.Firstly，the authors calculate fuzzy indiscernibility values between objects in each attribute based on fuzzy similarity measure.With these values the authors can construct fuzzy equivalence relations among objects.Then the authors can calculate the information entropy of any attribute set.The entropy-based attribute reduction algorithm can be used to acquire rules.Finally，by an example，the approach is verified to be reasonable and effective.

Key words: fuzzy rough set, fuzzy similarity measure, fuzzy equivalence relations, information entropy