NPCCPM: An Improved Approach Towards Community Detection in Online Social Networks
In this paper, we focus on the task community detection in social networks as it is the key aspect of complex network analysis. Lot of work has already been carried out on community detection, though most of the work done in this field is on non-overlapping communities. But in real networks, some nodes may belong to more than one community, so overlapping community detection needs more attention. The most popular technique for detecting overlapping communities is the Clique Percolation Method (CPM) which is based on the concept that the internal edges of a community are likely to form cliques due to their high density. CPM uses the term k-clique to indicate a complete sub-graph with k vertices. But it is not clear a priori which value of k one has to choose to detect the meaningful structures. Here we propose a method NO PARAMETER CORE CPM (NPCCPM) which calculates the value of k dynamically. Dynamic calculation of k makes it sure to give out the good community structure. We have developed a tool that improves the quality of simple CPM by making CPM-cover much more efficient by absorbing all the eligible nodes to communities and leaving out the bad nodes as outliers with respect to the given new detected cover.
KeywordsSocial networks Community detection Clique CPM NPCCPM Overlapping communities
- 1.Fortunato, S.: Community detection in graphs. Physics reports (2010)Google Scholar
- 4.Yang, J., Leskovec, J.: Defining and evaluating network communities based on ground-truth. In: ICDM 2012 (2012)Google Scholar
- 5.Granovetter, M.S.: The strength of weak ties. Am. J. Sociol. (1973)Google Scholar
- 11.Raghavan, U.N., Albert, R., Kumara, S.: Near linear time algorithm to detect community structures in large-scale networks. Phys. Rev. E Stat. Nonlin. Soft Matter Phys. 76(3), 036106 (2007)Google Scholar
- 12.Subelj, S., Bajec M.: Unfolding communities in large complex networks: combining defensive and offensive label propagation for core extraction. Phys. Rev. E Stat. Nonlin. Soft Matter Phys. 83(3), 036103 (2011)Google Scholar