Leader Similarity Based Community Detection Approach for Social Networks

In this era, problem of community detection becomes a fundamental provocation in social network to understand recognize the human relationship within network. Various researchers found methods of community detection and leader selection separately but our algorithm is working to find both community as well as leader of community appropriately. Human have tendency to connect with the people having same behavior or interest and the form community inside the network on the basis of common thoughts. Very few number of people present in the society who are responsible for the circulation of consciousness and represent a group of people who are overwhelmed by their thoughts and influenced by their information sharing phenomenon and treated as leader. In this paper, we propose a method named as ‘Leader Similarity Based Community Detection LSBCD’, this approach having mainly three steps. The first step is to select the leader node from the complete nodes of the network. The second step is to find similarity of nodes with the leader node. And in the third step we have to add isolated nodes of network into detected community. This approach deliver efficient results compare to ground truth communities.