An Efficient Algorithm to Find Influential Nodes Using Cliques in Social Networks

Influence maximization, a pivotal endeavor in social network analysis, seeks to discover a set of influential users capable of triggering dissemination of information, behaviors, or opinions throughout the network. Traditional approaches often rely on centrality metrics, like Degree or Closeness centrality, to determine influential nodes. However, these methods may overlook the importance of local network structures that can significantly impact information diffusion. In this manuscript, we introduce a new approach, CC-C, for influence maximization that leverages clustering coefficient and cliques to identify highly influencing users in social networks. The clustering coefficient is used to discover highly interconnected regions and potential influencers. Subsequently, cliques within the network are detected to locate densely interconnected subgroups. Integrating the clustering coefficient and clique information empowers the creation of a comprehensive influence ranking. To estimate the efficacy of proposed approach, experiments are performed utilizing the Independent Cascade (IC) model using real-world datasets. The experimental results reveal that proposed method, CC-C, outperforms traditional centrality-based methods in identifying influential nodes and their influence spread.