A Local-Global Influence Indicator Based Constrained Evolutionary Algorithm for Budgeted Influence Maximization in Social Networks

Abstract

Given a fixed total budget and a predefined cost model, the budgeted influence maximization problem aims to find a subset of nodes to maximize the influence spread in social networks while its cost should be no more than the fixed total budget. In this paper, we propose a local-global influence indicator based constrained evolutionary algorithm, named IICEA, to solve the budgeted influence maximization problem effectively and efficiently. In IICEA, a novel influence indicator is firstly designed by considering two components: local neighbor information and global community information, which can be used to better measure the influence of nodes in social networks. Based on the proposed local-global influence indicator, we propose a constrained evolutionary framework by designing several novel strategies such as mutation strategy, crossover strategy and repairment strategy to promote the evolution of population. Experimental results on 10 real-world social networks demonstrate the effectiveness of the proposed local-global influence indicator and also verify the effectiveness and efficiency of the proposed algorithm IICEA in comparison with several variant versions of IICEA and several representative baseline algorithms.