Predicting Node Influence in Complex Networks by the K-Shell Entropy and Degree Centrality

Currently, with the expansion of the use of social networks, the topic of information dissemination has achieved significant importance. The spread of rumors and efforts to stop them have led researchers to pay more attention than ever to predicting the impact of each individual on the network. Various methods have been proposed for this purpose, such as the Hybrid global structure model (HGSM), Generalized Gravity centrality (GGC), and Degree and neighborhood centrality (DNC). However, alongside their advantages, they have drawbacks, such as high time complexity, low accuracy, or inefficiency in distinguishing between the dissemination abilities of different individuals. Therefore, this paper focuses on a method based on degree, K-shell, and K-shell diversity in the neighborhood of each individual. Simulations were conducted using the Susceptible-Infected-Recovered (SIR) model and compared with 9 recent methods. Evaluations of 7 different networks in terms of resolution, accuracy, time complexity, and correlation exhibit the superiority of the proposed method.