A Survey on the Identification of Influential Spreaders in Complex Networks

Complex networks are useful for modeling many real-life complicated phenomena, including social networks. The social network has taken on a variety of identities, providing a wide range of services: virtual networking sites (e.g., Facebook, LinkedIn), microblogs, community media sites, etc. In complex networks, detecting the most influential spreaders is a crucial task. The traditional centrality methods are degree centrality, betweenness centrality, closeness centrality, page rank, and Katz centrality measures. In top-k ranking mode, these measures can determine the individuals most responsible for spreading an infection within a complex network, but there are still some issues. Our study examines a variety of existing global centrality measures, including k-shell decomposition and k-core, as well as their merits and demerits.