Identifying Top-N Influential Nodes in Large Complex Networks Using Network Structure

Abstract

Online social networks are popular for various activities like spreading information, creativity, and ideas, especially for viral marketing. The main focus in the social influence analysis, known as the influence maximization problem (IMP), aims to select top-N nodes to maximize the expected number of nodes activated by the top-N nodes (a.k.a seed nodes). This issue has gotten a lot of attention and has been looking into the issue of IMP, and these studies are usually too time-consuming to be useful in a complex social media network. The problem of seed selection is NP-hard. Due to the utilization of time-consuming Monte Carlo simulations, which are confined to small networks, a greedy method to the IMP issue is insufficient. The greedy approach, on the other hand, offers a good approximation assurance. In this paper, we present an algorithm for identifying communities and computing the ranking scores of nodes in the identified communities to solve the IMP with a focus on time efficiency.