An interval type-2 fuzzy PCM algorithm for pattern recognition

Abstract

The Possibilistic C-means (PCM) was proposed to overcome some of the drawbacks associated with the Fuzzy C-means (FCM) such as improved performance for noise data. However, PCM possesses some drawbacks such as sensitivity in the initial parameter values and to patterns that have relatively short distances between the prototypes. To overcome theses drawbacks, we propose an interval type-2 fuzzy approach to PCM by considering uncertainty in the fuzzy parameter m in the PCM algorithm. ©2009 IEEE.