摘要
在模糊信息系统中,通常用模糊相容关系代替等价关系.在已有模糊相容关系的基础上,通过利用有序加权平均(OWA)算子聚合所有对象之间在属性集中每个属性下的差异性,从而得出了对象间在整个属性集下的一个区分度,进一步地根据区分度定义了对象的相似度,给出了一种基于对象相似度的模糊相容关系.然后,对能影响该模糊相容关系的主要两个因素:阈值和模糊量词参数进行了具体分析.最后,通过实例结果说明在模糊信息系统中,这种改进的基于对象相似度的模糊相容关系能更灵活合理的处理问题.
In the fuzzy information system, the equivalence relation is instead of the fuzzy tolerance relation usually. Based on the existing fuzzy tolerance relation, the discrimination degree of objects is given by using the ordered weighted averaging （ OWA） operator, which can aggregate discrimination of every attribute in the attribute set for all the objects. Then the similar degree of objects is defined. A new fuzzy tolerance relation based on the similar degree of objects is presented. Two factors can influence the establishment of the fuzzy tolerance relation. Therefore, two factors and its influence are analyzed in detail. Finally, the experimental results show that the improved fuzzy tolerance relation is built according to similar degree of objects, which is more agile and reasonable.
出处
《小型微型计算机系统》
CSCD
北大核心
2014年第9期2131-2135,共5页
Journal of Chinese Computer Systems
基金
四川省教育厅科研项目(11ZB068)资助
四川师范大学重点研究课题项目资助
关键词
模糊信息系统
粗糙集
相似度
模糊相容关系
OWA算子
fuzzy information system
rough set
similar degree
fuzzy tolerance relation
OWA operator
