Kernel-based modified fuzzy possibilistic c-means clustering

El Sbai, Hassan
Classification and clustering algorithms are, without doubt, a useful tool to explore data structures, and have been widely employed in many domains such as pattern recognition, image processing, data mining, and data analysis. The focus of this paper is the partitioning problem with a special interest in kernel method. The aim of this paper is to extend this method to the modified fuzzy possibilistic c-means (MFPCM) algorithm. It is realized by substitution of a kernel-induced distance metric
more » ... or the Euclidean distance, and the corresponding algorithm is called kernel MFPCM algorithm. Numerical simulations are given to illustrate the performances of the proposed method.