AN APPROACH TO REMOVE THE EFFECT OF RANDOM INITIALIZATION FROM FUZZY C-MEANS CLUSTERING TECHNIQUE

Abstract

Out of the different available fuzzy clustering techniques Bezdek’s Fuzzy C-Means clustering technique is among the most popular ones. Due to the random initialization of the membership values the performance of Fuzzy C-Means clustering technique varies significantly in its different executions. We have tried to remove the effect of random initialization from Fuzzy CMeans clustering technique by using the Subtractive clustering technique of Chiu as a preprocessor to it. We have also provided a comparison of the performance of our method with those of Fuzzy C-Means clustering technique and Subtractive clustering technique.

Keywords

• Fuzzy C-Means Clustering Technique
• Subtractive Clustering Technique