Identifying influential spreaders in complex networks based on network embedding and node local centrality

Abstract

Identifying influential spreads in a network is of great significance for the analysis and control of the information dissemination process in complex networks. Based on the network embedding method, we propose an algorithm to identify the high influence nodes of the network. Firstly, the DeepWalk network embedding algorithm is used to map the high-dimensional complex network to a low-dimensional vector space to calculate the Euclidean distance between the local node pairs. Then, combined with the network topology information, a local centrality index of the network nodes is proposed to identify the high influence nodes. In eight real networks, the new algorithm is compared with five well-known identification methods. Numerical simulation results show that the new algorithm has a good performance in identifying influential spreaders.