Node importance ranking method in complex network based on gravity method

Abstract

How to use quantitative analysis methods to identify which nodes are the most important in complex network, or to evaluate the importance of a node relative to one or more other nodes, is one of the hot issues in network science research. Now, a variety of effective models have been proposed to identify important nodes in complex network. Among them, the gravity model regards the coreness of nodes as the mass of object, the shortest distance between nodes as the distance between objects, and comprehensively considers the local information of nodes and path information to identify influential nodes. However, only the coreness is used to represente the quality of the object, and the factors considered are relatively simple. At the same time, some studies have shown that the network can easily identify the core-like group nodes with locally and highly clustering characteristics as core nodes when performing <i>k</i>-core decomposition, which leads to the inaccuracy of the gravity algorithm. Based on the universal gravitation method, considering the node <i>H</i> index, the number of node cores and the location of node structural holes, this paper proposes an improved algorithm ISM and its extended algorithm ISM<sub>+</sub>. The SIR model is used to simulate the propagation process in several classical real networks and artificial networks, and the results show that the proposed algorithm can better identify important nodes in the network than other centrality indicators.