Selecting Parameters of the Fuzzy Possibilistic Clustering Algorithm

Abstract

Clustering has been widely used in pattern recognition, image processing, and data analysis. It aims to organize a collection of data items into clusters, such that items within a cluster are more similar to each other than they are in other clusters. The Fuzzy Possibilistic C-Means (FPCM) is one of the most popular clustering methods based on minimization of a criterion function. So the implementation of this algorithm requires a priori selection of some parameters: the fuzzy and the typical exponent, initialization of cluster centers. But the definition of these parameters at the moment is fixed in advanced and the initialization of centers is random; so the algorithm can give results not consistent. The determination of an optimal value for these parameters and the cluster centers at the beginning are problematic and remains an open problem. New procedures for choice of the optimal values of parameters and for initialization of centers were developed. Numerical results using data sets are used to illustrate the simplicity and effectiveness of the proposed procedures. General Terms Pattern Recognition, Clustering, Algorithms et. al.