Lexical sorting centrality to distinguish spreading abilities of nodes in complex networks under the Susceptible-Infectious-Recovered (SIR) model

Epidemic modeling in complex networks is a hot research topic in recent years. The spreading of a virus (such as SARS-CoV-2) in a community, spreading computer viruses in communication networks, or spreading gossip on a social network is the subject of epidemic modeling. The Susceptible-Infectious-Recovered (SIR) is one of the most popular epidemic models. One crucial issue in epidemic modeling is the determination of the spreading ability of the nodes. Thus, for example, super spreaders can be detected in the early stages. However, the SIR is a stochastic model, and it needs heavy Monte-Carlo simulations. Hence, the researchers focused on combining several centrality measures to distinguish the spreading capabilities of nodes. In this study, we proposed a new method called Lexical Sorting Centrality (LSC), which combines multiple centrality measures. The LSC uses a sorting mechanism similar to lexical sorting to combine various centrality measures for ranking nodes. We conducted experiments on six datasets using SIR to evaluate the performance of LSC and compared LSC with degree centrality (DC), eigenvector centrality (EC), closeness centrality (CC), betweenness centrality (BC), and Gravitational Centrality (GC). Experimental results show that LSC distinguishes the spreading ability of nodes more accurately, more decisively, and faster.