Influential spreaders identification in complex networks with potential edge weight based k-shell degree neighborhood method

In social network analysis, one of the important aspects is to identify the nodes that are vital to the information diffusion process such as viral marketing, worm propagation in a computer network, disease spreading, etc. The degree centrality measures only the local influence of a node whereas, others like closeness and betweenness capture the global impact. K-shell centrality uses the position/location of a node in the network to estimate its spreading ability. Most of the above techniques except degree perform well when the network is complete. The degree centrality does not require complete network information but fails to identify many important nodes due to limited use of only local information. In this paper, we propose a new measure namely potential edge weight based k-shell degree neighborhood centrality to rank the node's spreading ability without depending on the degree of completeness of the network. The proposed method uses node degree and k-shell index along with a derived network parameter to assign potential edge weights to the connecting links between two nodes. Information propagation is simulated using Susceptible-Infected-Recovered (SIR) epidemic model and performance comparison of the proposed method is done using Kendall's rank correlation with other ranking techniques. Experiments on real network establish the superiority of the proposed method in identifying influential spreaders in comparison to a degree, k-shell and other standard ranking techniques. Computationally the proposed method is cost-effective even with large complex networks. (C) 2019 Elsevier B.V. All rights reserved.