MODIFIED POSSIBILISTIC FUZZY C-MEANS ALGORITHM FOR CLUSTERING INCOMPLETE DATA SETS

A possibilistic fuzzy c-means (PFCM) algorithm is a reliable proposed to deal with the weaknesses associated handling noise sensitivity and coincidence clusters in (FCM) (PCM). However, PFCM only applicable complete data sets. Therefore, this research modified for clustering incomplete sets OCSPFCM NPSPFCM performance evaluated based on three aspects, 1) accuracy percentage, 2) number of iterations, 3) centroid errors. The results showed that outperforms missing values ranging from 5% ? 30% all experimental Furthermore, both algorithms provide average accuracies between 97.75%?78.98% 98.86%?92.49%, respectively.