An Improved Validity Function for Fuzzy C-Means Cluster
Yibing Wu, Song Jianshe, Chao Niu · 2011
Fuzzy c-mean (FCM) is an algorithm for obtaining an optimal fuzzy partition of data set by minimizing an objective function. The cluster validity function is used to evaluate the validity of clustering, and the clustering results will tend to be more reasonable on the condition that the initial clustering number is accurately ascertained. According to the analysis of the weighting exponent m, a new cluster validity function is proposed based on the intra-cluster disperse distance. Then the stability and reliability of the function is analyzed theoretically. The experimental results indicate that the new validity function can find out the optimized cluster number and it is also robust to the weighting coefficient m.