How opinion distortion appears in super-influencer dominated social network

Abstract Social network platforms are rather popular nowadays. Since everyone can express their opinions on social platforms without a gatekeeper, opinion bias, polarization, and distortion have been increasingly commonplace. In this paper, we focus on the opinion distortion triggered by the super-influencer who is more powerful and influential than general opinion leaders. Considering three features of the super-influencer, and the controversial information posted by the super-influencer that will cause an intense discussion on social platforms, we constructed an extended model for simulation of the opinion evolution on social platforms. Then, we performed several experiments using MATLAB, considering the influence of three factors including authority, evolutionary time, and arguable information by adjusting the parameter values. Results showed that, compared with the situation without a super-influencer, public opinion could be significantly distorted when there existing a super-influencer, as long as the evolutionary time is sufficient, and the overall network opinion would be inclined to the super-influencer. Although arguable information can disturb the convergence process of opinions, the average opinion value still polymerizes to a favored opinion interval.