Ranking the spreading capability of nodes in complex networks based on link significance

Evaluating the spreading capability of nodes in complex networks is highly significant for understanding the spreading behavior in complex networks. Recent studies showed that the degree centrality and the coreness centrality can effectively evaluate the spreading capability of nodes. However, specific network topologies significantly decrease the effectiveness of these two metrics. In this paper, we propose a new method called LS method. The LS method distinguishes the importance of the different edges of the node during the spreading process and figures out a new way to rank the spreading capability of nodes. Simulation experiments on real networks show that the proposed method is more efficient and more versatile than the degree centrality and the k-shell decomposition algorithms. Unlike other improved methods, the LS method does not need to compute any empirical parameter, which means the LS method is more efficient than these improved methods. (C) 2018 Elsevier B.V. All rights reserved.