Exploring Community Structures for Influence Maximization in Social Networks

Since the surge of the popularity of social network, recently, there has been a tremendous wave of interest in the investigation of influence maximization problem. Given a social network structure, the problem of influence maximization is to determine a minimum set of nodes that could maximize the spread of influences. Nowadays, due to the dramatic size growing of social network, the efficiency and scalability of algorithms for influence maximization become more and more crucial. Although many recent studies have focused on the problem of influence maximization, these works, in general, are time consuming when a large-scale social network is given. In this paper, by utilizing community structures, we develop two efficient algorithms CDHKcut and CDH-SHRINK (standing for Community and Degree Heuristic with Kcut / SHRINK) that significantly decrease the number of candidate influential nodes. The experimental results on real datasets indicate that our algorithms not only significantly outperform state-of-the-art algorithms in efficiency but also possess graceful scalability.