An SEI 3 R information propagation control algorithm with structural hole and high influential infected nodes in social networks
Qian Zhang,Xianyong Li,Yongquan Fan,Yajun Du
DOI: https://doi.org/10.1016/j.engappai.2021.104573
IF: 8
2022-02-01
Engineering Applications of Artificial Intelligence
Abstract:Information propagation and control have great significance to manage public opinion in social networks. This paper aims to establish a novel information propagation model and the corresponding control algorithm to display the processes of information evolution, propagation and control. Considering high influential nodes and structural hole nodes highly influence the decision-making choices of public users’ opinions in the information propagation process, the structural hole node discovery and score (SHNDS) algorithm and high influential node discovery and score (HINDS) algorithm are proposed, respectively. According to different node status for a social network, the network nodes are divided into susceptible ( S ), latent ( E ), ordinary influential infected ( I n ), structural hole infected ( I s ), high influential infected ( I h ) and removed nodes ( R ). Then, combining with the SHNDS and HINDS algorithms, a new SEI 3 R information propagation model and the corresponding control algorithm are proposed. Experimental results show that when the high influential nodes are regarded as the initial information propagation nodes, the information propagates to all nodes of their communities which contained the initial information propagation nodes. The information propagation ranges are wider than SEIR and SEI 2 R models. When the structural hole nodes are taken as initial information propagation nodes, the information propagates to the entire networks. When the structural hole nodes and high influential nodes are both seen as the initial information propagation nodes, information propagation speeds and ranges are faster and wider than the former two cases, respectively. By controlling high influential nodes, structural hole nodes, and both of them, the numbers of nodes received the information are reduced in turn. These indicate that information control effects are gradually improved.
automation & control systems,computer science, artificial intelligence,engineering, electrical & electronic, multidisciplinary