Information entropy based public opinion maximization in social networks

Abstract

Aiming at addressing the public opinion maximization problem in social networks with more intelligence, we propose an information entropy-based method. First of all, considering the different information carried by different types of social network nodes and the different information transmitted by different social nodes, the definitions of participation entropy and interactive entropy are proposed. Then, the influence weight between public opinion propagation nodes is calculated, and then the global influence of nodes is calculated based on the linear threshold model. Finally, the seed set is selected according to the marginal gain of the social nodes. The experimental results show that the proposed algorithm outperforms the other state-of-the-art methods.