Optimized initial centers for fuzzy C-means algorithm

The fuzzy C-means(FCM) algorithm is sensitive to initial clustering centers.This paper introduces an initial center selecting algorithm based on the weight through computing the sample weight and selecting representative samples as initial clustering centers.The optimized initial centers are presented for the FCM algorithm.In comparison with traditional algorithms,the presented algorithm can get more steady cluster results,and the clustering accuracy is also improved.Experiment shows the effectiveness of the improved algorithm.