Novel intuitionistic fuzzy c-means clustering for linearly and nonlinearly separable data

This paper presents a robust Intuitionistic Fuzzy c-means (IFCM-σ) in the data space and a robust kernel Intutitionistic Fuzzy C-means (KIFCM-σ) algorithm in the high-dimensional feature space with a new distance metric to improve the performance of Intuitionistic Fuzzy C-means (IFCM) which is based upon intuitionistic fuzzy set theory. IFCM considered an uncertainty parameter called hesitation degree and incorporated a new objective function which is based upon intutionistic fuzzy entropy in the conventional Fuzzy C-means. It has shown better performance than conventional Fuzzy C-Means. We tried to further improve the performance of IFCM by incorporating a new distance measure which has also considered the distance variation within a cluster to regularize the distance between a data point and the cluster centroid. Experiments are done using two-dimensional synthetic data-sets, Standard data-sets referred from previous papers. Results have shown that proposed algorithms, especially KIFCM-σ is more effective for linear and nonlinear separation.