Published By: Blue Eyes Intelligence Engineering

Abstract — Clustering is a collection of objects which are similar between them and dissimilar to the objects. Three steps to form cluster is to initialize data, select data randomly and use distance metrics to form optimal solution. In this paper fuzzy c-means algorithm is proposed. It is a kind of partitional clustering algorithm because it partition the data sets into several subsets. It allows one part of data in which it belong to two or more cluster. Data sets are taken as input value. There are many data sets so randomly select the variables to form the cluster group. It gives more efficient and effective clustering. Index Terms—Clustering, Collective fuzzy c-means, K-harmonic mean