Title: Finding high-influence microblog users with an improved PSO algorithm

Authors: Biao Zhang; Shuai Zhong; Kunmei Wen; Ruixuan Li; Xiwu Gu

Addresses: School of Mechanical Engineering and Automation, Wuhan Textile University, Wuhan, 430074, China ' School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, 430074, China ' School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, 430074, China ' School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, 430074, China ' School of Computer Science and Technology, Huazhong University of Science and Technology, Wuhan, 430074, China

Abstract: Particle swarm optimisation (PSO) is a stochastic optimisation algorithm based on swarm intelligence. The algorithm applies the concept of social interaction to find optimal solution. Sina Weibo is one of the most popular Chinese microblog platforms. Microblog users participate in network interaction by publishing tweets and retweets. The influences of microblog users are determined by the users' behaviours, which exactly match the five principles of swarm intelligence. Therefore, we propose an improved PSO algorithm to find the microblog users with the maximum influence. Microblog users' retweeting behaviours can be described as a variable of the user influence space, which contains user experiences and surrounding network. The variable is defined as the velocity change in our method. By iteratively calculating based on users' behaviour, the maximum influence will be obtained. The experiments validate that our method can effectively identify the high-influence microblog users.

Keywords: particle swarm optimisation; PSO; social networks; Sina Weibo; microblogs; high-influence microblog users; swarm intelligence; tweets; retweets; maximum influence; retweeting behaviour.

DOI: 10.1504/IJMIC.2013.053540

International Journal of Modelling, Identification and Control, 2013 Vol.18 No.4, pp.349 - 356

Published online: 29 Apr 2013 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article