A novel neighborhood-based importance measure for social network influence maximization using NSGA-III

Information spreading on social networks is a chain process that depends on various factors for its speed of spread. Influential users with an impact on many users play a significant role in information dissemination on the network. Finding influential users, which is known as maximizing influence, can be useful for many applications in social network analysis. Recently, metaheuristic algorithms have been used as tools for finding influential users based on social importance criteria to solve the problem of maximizing influence. In this article, a method for maximizing influence based on the non-dominated sorting genetic algorithm version III (NSGA-III) using a novel similarity importance criterion is proposed. The proposed method has been tested under various scenarios with a limited number of influential users on the Facebook dataset. The results of the experiments showed that the proposed method was able to identify influential users more optimally using the similarity criterion, improving the number of users under influence in social networks by $ 9.5%,\ 16.5%$, and $ 6.5%$ compared to centrality, closeness, and betweenness criteria, respectively.