A Genetic NewGreedy Algorithm for Influence Maximization in Social Network

A user may be influenced by the other users of a social network by sharing information. Influence maximization is one of the critical research topics aimed at knowing the current circumstances of a social network, such as the general mood of the society. The goal of this problem is to find a seed set which has a maximum influence with respect to a propagation model. For the influence maximization problem is NP-Hard, it is obvious that an exhausted search algorithm is not able to find the solution in a reasonable time. It is also obvious that a greedy algorithm may not find a solution that satisfies all the requirements. Hence, a high-performance algorithm for solving the influence maximization problem, which leverages the strength of the greedy method and the genetic algorithm (GA) is presented in this paper. Experimental results show that the proposed algorithm can provide a better result than simple GA by about 10% in terms of the quality.