On rough fuzzy cluster validity indices

This study analyses the cluster validity measures used for rough set based clustering algorithms. Cluster validity measures have been used to evaluate the quality of cluster partitions. Rough fuzzy C means clustering uses the concept of crisp lower approximation and fuzzy boundary. Cluster evaluation is the final step in clustering process. Even though rough fuzzy clustering is less descriptive than fuzzy clustering but more descriptive than rough clustering, the resultant rough fuzzy clusters must be evaluated in order to obtain good clustering results. This paper presents some of the popular rough fuzzy cluster quality measures used to evaluate the performance of rough fuzzy clustering algorithm. It also evaluates Rough clustering algorithm using those measures and the results shows that these measures provides accurate results for the Rough set based clustering algorithms. It additionally proves that RFCM provides good results than RCM.