A new validity index for fuzzy-possibilistic c-means clustering

Abstract

In some complicated datasets, due to the existence of noisy data points and outliers, cluster validity indices can yield conflicting results in terms of determining the optimal number of clusters. This paper presents a new validity index for fuzzy-possibilistic C-means clustering called Fuzzy-Possibilistic (FP) index, which works well in the presence of clusters that vary in shape and density. Moreover, like most of the clustering algorithms, Fuzzy-Possibilistic C-Means (FPCM) is susceptible to some initial parameters. In this regard, in addition to the number of clusters, F P C M requires a priori selection of the degree of fuzziness (m) and the degree of typicality (77). Therefore, an efficient procedure was presented for determining optimal values of m and n. The proposed approach is evaluated using several synthetic and real-world datasets. Final computational results demonstrate the capabilities and reliability of the proposed approach compared with several well-known fuzzy validity indices in the literature. Furthermore, to clarify the ability of the proposed method in real applications, the proposed method is implemented in microarray gene expression data clustering and medical image segmentation.