Enhanced MCL Clustering

Abstract

The goal of graph clustering is to partition vertices in a large graph into different clustersbased on various criteria such as vertex connectivity or neighborhood similarity. Graphlustering techniques are very useful for detecting densely connected groups in a large graph. Inthis research, we introduce a clustering algorithm for graphs; this algorithm is based on Markovlustering (MCL), which is a clustering method that uses a simulation of stochastic flow. Wehave tuned to set the proper factors of inflation, matrix and threshold. Theoretical analysis isprovided to show that the enhanced EMCL-Cluster is converging. Then the proposed method isompared with other clustering methods.