A Novel Hybrid Fuzzy Clustering Algorithm Based on Artificial Bee colony and C-means

In this study a novel hybrid clustering algorithm is proposed that is based on Artificial Bee Colony (ABC) and Fuzzy C-means (FCM). When FCM is applied on high dimensional dataset, it usually results in local optimal partitioning. In this paper we address this problem and used a recently developed evolutionary technique named Artificial Bee Colony in combination to FCM. Hence, the name is Fuzzy C-means Bee (FCB) algorithm. The method can detect globally optimal cluster centeroids better than FCM as a most wildly used and popular clustering technique. To demonstrate performance of the proposed algorithm of FCM we used it for some standard dataset UCI datasets. The results show that FCB converge to global faster than FCM and ABC

A Novel Hybrid Fuzzy Clustering Algorithm Based on Artificial Bee colony and C-means Keywords:

Fuzzy C-means , Artificial Bee Colony , Optimization , Clustering technique