Multiple kernel interval type-2 fuzzy c-means clustering

Abstract

In this paper, kernel interval type-2 fuzzy c-means clustering (KIT2FCM) and multiple kernel interval type 2 fuzzy c-means clustering (MKIT2FCM) are proposed as a base for classification problems. Besides building algorithms KIT2FCM to overcome some drawback of the conventional FCM and use the advantages of fuzzy clustering technique on the interval type 2 fuzzy set in handling uncertainty, the paper also introduces combining the different kernels to construct the MKIT2FCM which provides us a new flexible vehicle to fuse different data information in the classification problems. That is, different information represented by different kernels is combined in the kernel space to produce a new kernel. The experiments are done based on well-known data-sets and application of land cover classification from multi-spectral with the statistics show that the algorithms generates good quality of classifications.