Automatic determination of the number of clusters is a very important issue in cluster analysis. In this paper, we explore Fuzzy C-Means (FCM) based clustering algorithms to determine the number of clusters in a data set through cluster validity optimization. To improve the computation speed, we propose two strategies for eliminating and for spliting a cluster allowing the FCM-based algorithms to make efficient use of cluster centers computed at each step. To improve existing validity measures, we make use of a new validity function that fits particularly data sets containing overlapping clusters. Experimental results will be given to illustrate the performance of the new algorithms.
展开▼