A New Validity Index for Fuzzy-Possibilistic C-Means Clustering

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