A Novel Algorithm for Automatically Detecting Number of Clusters for Mining Communities in Heterogeneous Social Networks
Renuga Devi R1, Hemalatha. M2
1Renuga Devi R, Department of Computer Science, Karpagam University, Coimbatore (Tamil Nadu), India.
2Hemalatha M, Department of Computer Science, Karpagam University, Coimbatore (Tamil Nadu), India.
Manuscript received on 10 November 2013 | Revised Manuscript received on 18 November 2013 | Manuscript Published on 30 November 2013 | PP: 43-47 | Volume-3 Issue-6, November 2013 | Retrieval Number: F1341113613/13©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Social media have attracted millions of user’s attention in recent years. In a distributed social network a community mining is one of the major research areas. Mining of network communities is a major problem now a day. This problem should be avoided. Several methods were proposed, but most of the methods of community mining consider the homogeneous network. But in distributed network there are multiple networks are interconnected with each other which are known as heterogeneous networks. Each network represents a specific kind of relationship. Same time each relationship plays an important place in a distinct situation. Mining of such an important community in a distributed environment is a difficult task. To overcome the above mentioned problem, this paper presents a novel Convergence aware Dirichlet Process Mixture Model (CADPM) for automatically mining the network communities in heterogeneous networks. The earlier Dirichlet Process (DP) mixture model is unsuitable in some situation. The number of clusters for community mining is unknown in prior. So the CADPM is proposed to handle the large number of data-cases.
Keywords: Community, Dirichlet Process, Heterogeneous Network, Hidden Communities, Social Network.
Scope of the Article: Social Networks