Global Fuzzy C-Means with Kernels

Global Fuzzy C-Means with Kernels
Authors : Gyeongyong Heo, Hun Choi, Jihong Kim
Publication Date: 05-12-2016


Author(s):  Gyeongyong Heo, Hun Choi, Jihong Kim

Published in:   International Journal of Engineering Research & Technology

License:  This work is licensed under a Creative Commons Attribution 4.0 International License.

Website: www.ijert.org

Volume/Issue:   Volume. 5 - Issue. 12 , December - 2016

e-ISSN:   2278-0181

 DOI:  http://dx.doi.org/10.17577/IJERTV5IS120090


Fuzzy c-means (FCM) is a simple but powerful clustering method using the concept of fuzzy sets that have been proved to be useful in many areas. There are, however, several well-known problems with FCM, such as sensitivity to initialization, sensitivity to outliers, and limitation to convex clusters. In this paper, a new clustering method, which is an extension of global fuzzy c-means (G-FCM) and kernel fuzzy c-means (K-FCM), is proposed to resolve the shortcomings mentioned above. G-FCM is a variant of FCM that uses an incremental seed selection method and is effective in alleviating the sensitivity to initialization. There are several approaches to reduce the influence of noise and properly partition non-convex clusters and K-FCM is one of them. K-FCM is used in this paper because it can easily be extended with different kernels, which provide sufficient flexibility to allow for resolution of the shortcomings of FCM. By combining G-FCM and K-FCM, the proposed method, kernelized global FCM (KG-FCM), can resolve the shortcomings mentioned above. The usefulness of KG-FCM is demonstrated by experiments using artificial and real world data sets.


Number of Citations for this article:  Data not Available


Key Word(s):    


Number of Downloads:     41

Call for Papers - May - 2017



                 Call for Thesis - 2017 

     Publish your Ph.D/Master's Thesis Online

              Publish Ph.D Master Thesis Online as Book