A local rewiring strategy for accelerating information propagation under complex propagation mechanism

Abstract

Abstract Controlling the spread of information or rumors in complex networks has always been a subject of extensive attention and research. Generally speaking, taking the most influential node in the network as the propagation spreader or cutting off the most important link in the propagation process can maximize the promotion or suppression of information transmission. Therefore, many scholars have put forward many measurement methods to identify specific nodes and links in the network, and their theories are becoming more and more mature. However, in the face of large-scale networks, most of these methods are exposed to high computational complexity, low accuracy, and only applicable to specific networks. Some scholars try to enhance (inhibit) the dissemination of information by adding or deleting links, but the structural characteristics of the original network will be changed to a large extent, which also means that some basic functions related to the network have been changed. In this work, we propose a rewiring strategy to accelerate information propagation under complex propagation mechanism, which increases the number of redundant links by rewiring the second-order neighborhood of the initial propagation node. At the same time, the rewiring strategy preserves the degree distribution of nodes and modularization feature of the original network to a large extent. We apply our strategy to simulate the complex propagation of the information propagation (IST(I)) model on different networks. The simulation results show that the rewired network can accelerate the propagation of information more effectively, and the number of spreaders in the network increases significantly, and the propagation rate of information is enhanced significantly. In addition, this paper also proposed for the first time an index that quantifies both the transmission ability and infectivity of the information at the micro level, namely the number of regenerated infections κ , which was optimally estimated by statistical inference. We also compare the changes of various topological characteristics of the network before and after the rewiring. We find that the structural characteristics related to the propagation dynamics of the network after the rewiring are significantly enhanced.