模糊c-均值聚类算法中加权指数m的研究

 加权指数m是模糊c-均值(FCM)聚类算法中的一个重要参数.本文从FCM算法出发研究了m对聚类分析的影响,m的最佳选取方法及其在聚类有效性中的应用三个问题.实验结果表明:m不合适的取值将严重影响算法的性能;在实际应用中m的最佳取值范围为 ,这与Pal的实验结论相一致;另外基于最优加权指数m^*的类别数确定方法是相当有效和灵敏的.

A Study of Weighting Exponent m in a Fuzzy c-Means Algorithm

Weighting exponent m is an important parameter in fuzzy c means(FCM) algorithm.In this paper,three basic pro blems are studied in FCM algorithm:the effect of m on the performance of fuzzy clustering,the optimal choice of m ,and its preliminary applications in clustering validity.Experimental results indicate that an improper choice of m will influence critically on the performance of clustering,and the optimal range of m is within 1 5,2 5] in the practical applications,which is consistent with the conclusion of Pal.In addition,the approach to determining the optimal number of clusters based on the optimal weighting exponent m * is much effective and sensitive.