【摘要】 研究了基于区间值模糊集上的模糊聚类.先讨论了基于∨-t关系合成基础上的一种n次幂运算,是Yang和Shih的n次幂运算的推广.通过运算,由模糊相似关系矩阵得到模糊等价关系矩阵,并提出了一种基于模糊等价矩阵的聚类算法.针对非完备模糊相似矩阵中的数据信息可能会缺失的情况,最后讨论了一种由∨-t合成运算来推算缺失值的简单方法.更多还原

【Abstract】 The fuzzy clustering is discussed based on intervalvalued fuzzy relations.On the basis of ∨-t compositions,we discuss an n-step procedure which is an extension of Yang and Shih’s n-step procedure.A similarity-relation matrix is obtained by beginning with a proximity-relation matrix using the proposed n-step procedure.Then we proposed a clustering algorithm for the similarity-relation matrix.We also discuss a simple form of imputation to estimate missing values by ∨-t compositions.更多还原