A Weight Possibilistic Fuzzy C-Means Clustering Algorithm

Fuzzy C-means (FCM) is an important clustering algorithm with broad applications such as retail market data analysis, network monitoring, web usage mining, and stock prediction. Especially, parameters in FCM have influence on results. However, a lot of did not solve the problem, that is, how to set parameters. In this study, we present kind method for computing values according role process. New are assigned membership typicality so modify objective function, basis which Lagrange equation constructed iterative acquired, does center equation. At last, new possibilistic fuzzy based weight parameter (WPFCM) was proposed. order test efficiency algorithm, some experiments different datasets conducted compare WPFCM FCM, (PCM), (PFCM). Experimental results show times less than about 25% PFCM 65% dataset X12. Resubstitution errors 19% PCM 74% 10% IRIS dataset.