基于模糊属性刻画的T相似性关系的构造  

Construction of T-similarity Relations Based on Fuzzy Attribute Descriptions

作　　者：唐敏[1] 杨宏文[1] 胡卫东[1] 郁文贤[2] 

机构地区：[1]国防科学技术大学ATR重点实验室,湖南长沙410073 [2]上海交通大学电子信息与电气工程学院,上海200030

出　　处：《模糊系统与数学》2011年第5期136-143,共8页Fuzzy Systems and Mathematics

基　　金：国家部委基金资助项目(513060302)

摘　　要：模糊关系是模糊粗糙分析的基础。从属性数据生成模糊关系是模糊粗糙集实际应用中的重要问题。针对模糊属性刻画,给出了生成几种T相似性关系的方法。首先,对于每一个属性,分别生成一个T相似性关系。而后,通过聚合算子来合成这些T相似性关系,以得到一个综合的T相似性关系。The data descriptions to objects are generally classified into two types,the object data,in which each object is described explicitly by a list of attributes,and the relational data,in which only pairwise similarity or dissimilarity are given.Object data can always be transformed into relational data by using certain distance or similarity function.Processing after transformation is a useful strategy in the case of mixed data types.Moreover,it is more convenient to analyze the resulting relational data when the attribute characterizations of object data are fuzzy.The construction of T-similarity relations from fuzzy attribute characterizations is considered in this study.We develop a scheme for systematically generating T-similarity relations for typical T-norms.Specifically,a partial T-similarity relation is derived for each attribute first,to which aggregation operators are applied to obtain a combined T-similarity relation.

关 键 词：模糊粗糙集 模糊属性数据 T相似性关系 相似性关系合成 

分 类 号：O519[理学—低温物理]