Extracting Influential Nodes in Social Networks Based On Community Structure

Abstract

Influence maximization (IM) is the process focuses on finding active users who make that maximizes the spread of influence into the network. In recent years, community detection has attracted intensive interest especially in the implementation of clustering algorithms in complex networks for community detection. In this paper the social network was divided into communities using the proposed algorithm which is called (CDBNN) algorithm, CDBNN stands for Community Detection Based on Nodes Neighbor. The key nodes (candidate nodes) were extracted using the degree centrality in each community. The propagates model (PSI) was used to information propagates through the network. Finally, using closeness centrality to extract the influential nodes from the network. Experimental results on the real network are efficient for influence propagates, compared with three known proposals.