Generalized Possibilistic C-Means Clustering Based on Differential Evolution Algorithm
Fuheng Qu, Siliang Ma, Yating Hu · 2009
In this paper, a new clustering model called generalized possibilistic c-means (GPCM) is proposed, and an efficient global optimization technique-differential evolution algorithm is used to optimize the proposed model. GPCM modifies possibilistic c-means (PCM) by limiting each cluster center in a fixed feasible region respectively. The feasible region is determined by the fuzzy c-means clustering algorithms, and then the optimal solution of GPCM model is searched by the differential evolution algorithm within the determined feasible region. GPCM inherits the noise robustness property of PCM, and it eliminates the coincident clusters problem of PCM by limiting different cluster centers in disjoint feasible regions. Experiments on the synthetic and real world data sets illustrate the effectiveness of GPCM.