Identifying multiple influential spreaders in term of the distance-based coloring

• We present an improved distance-based coloring method to identify the multiple influential spreaders.
• Each node is colored by a kind of color where the distance between initial nodes is close to the average distance.
• For three empirical networks show that the improvement ratio of our distance-based coloring method could reach 8.16% for the Erdos network.

Identifying influential nodes is of significance for understanding the dynamics of information diffusion process in complex networks. In this paper, we present an improved distance-based coloring method to identify the multiple influential spreaders. In our method, each node is colored by a kind of color with the rule that the distance between initial nodes is close to the average distance of a network. When all nodes are colored, nodes with the same color are sorted into an independent set. Then we choose the nodes at the top positions of the ranking list according to their centralities. The experimental results for an artificial network and three empirical networks show that, comparing with the performance of traditional coloring method, the improvement ratio of our distance-based coloring method could reach 12.82%, 8.16%, 4.45%, 2.93% for the ER, Erdős, Polblogs and Routers networks respectively.