摘要
针对现有的二型模糊系统规则库精简方法不能有效地消除冗余模糊集合的问题,提出了新的普通二型模糊相似度与包含度.首先,基于2种模糊性测度的公理化定义,提出了计算公式;然后,讨论了普通二型模糊包含度的性质,并提出了2种测度的相互转换定理;最后,通过实例验证了2种新测度的性能,并将提出的普通二型模糊相似度用于高斯普通二型模糊集合的聚类分析,聚类结果由不同α-水平上的层次聚类树组成,可以合理地区分这些集合.
To overcome the problem of simplified methods of rule bases of a type-2 fuzzy logic system can not eliminate the harmful effects of redundant fuzzy sets,a similarity measure and an inclusion measure between general type-2 fuzzy sets are proposed.Firstly,based on the selected axiomatic definitions of two fuzzy measures,the computation formulas are given by considering the footprint of uncertainty（FOU） and the secondary membership function.Then, several properties of the proposed inclusion measure are discussed,and the theorems that two fuzzy measures can be transformed by each other are presented.Finally,two examples are given to validate their performance and the proposed similarity measure is applied to clustering analysis of general type-2 fuzzy data. The cluster results consist of a hierarchical tree in different-levels and can reasonably differentiate these type-2 fuzzy sets.
出处
《华中科技大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
2011年第8期119-123,共5页
Journal of Huazhong University of Science and Technology(Natural Science Edition)
基金
国家自然科学基金资助项目(60674057)
中央高校基本科研业务费专项资金资助项目(SWJTU-09ZT11)
关键词
二型模糊集合
模糊性测度
相似度
包含度
聚类
type-2 fuzzy set
fuzzy measure
similarity measure
inclusion measure
clustering
