混合型模糊聚类分析方法及其应用

在动态聚类方法和模糊ISODATA方法的基础上,提出了混合型模糊聚类分析方法.该方法首先利用传统的传递闭包方法得到1个初始分类,并在此基础上提出初始分划矩阵,根据考虑权重因子的模糊ISODATA方法对相关数据进行迭代计算,从而对数据进行有效分类.以股票分类为例对该方法进行实证分析,分析结果表明,应用该方法可以对股票进行有效分类优选.

Mixed fuzzy clustering analysis and its application

By combination of the dynamic clustering method with ISODATA method,the mixed fuzzy clustering analysis method was put forward.According to the method,an initial classification was firstly performed by use of the conventional transitive closure method,and then an initial division matrix was constructed.With the weighted fuzzy ISODATA method for iterative computation of the related data,the data were classified effectively.As a case study,the method was applied to stock classification,and the result shows that the method is effective for stock classification and optimal selection.