Research On Microblog User Influence Based On Particle Swarm Optimization

As a new type of social network in Web2.0era, microblog becomes very popularbecause of its low threshold for participation, real time to publish information and sociability.Research on microblog user influence is useful for user to find some valuable informationsource, thereby improve user’s efficiency on filtering information. This research is alsoimportant on public dissemination and viral marketing.Most traditional measures of calculating user influence are based on the theory ofcomplex networks. There have some factors to determine user’s influence like userbehaviors and user relations.Particle swarm optimization (PSO) is a stochastic optimization algorithm based onswarm intelligence. The algorithm applies the concept of social interaction to find optimalsolution. Microblog Users participate in network interact by publishing tweets andretweeting. The influences of microblog users are determined by the users’ behaviors, whichexactly match the five principles of swarm intelligence.Therefore, this paper proposes an improved PSO algorithm to find the microblog userswith the maximum influence. Microblog users’ retweeting behaviors can be described as avariable of the user influence space, which contains user experiences and surroundingnetwork. The variable is defined as the velocity change in our method. By iterativelycalculating based on users’ behavior, the maximum influence will be obtained.The experiments validate that our method can effectively identify the high-influencemicroblog users.