Opinions Evolution Analysis in Time-varying Social Networks: A Silent Spiral Opinion Model

Abstract

The spiral of silence theory originating from the mass media era can no longer adapt to the phenomenon of social media era. Opinions in social networks are so diverse that people have opportunities to voice their opinions and prevent isolation. In this paper, we focus on an interesting phenomenon: the generation of pseudo-opinions caused by the spiral of silence and its impact on interpersonal relationships. We propose a silent spiral opinion model (SSOM) to explore how this phenomenon affects opinion evolution, as well as interpersonal relationships. We model the spiral of silence in which silent agents are given pseudo-opinions. In addition, we propose a time-varying network topology updating algorithm based on triangle closure to investigate how the spiral of silence would affect interpersonal relationships. Finally, we conduct simulations and analysis in ER random networks, which are easy to vary in size and density and capture some of the essential features of social networks. The experimental results reveal that the network size and the evolution of opinions are independent of each other, but when the network size increases, people are more likely to be silent and more distant from each other. The intensification of the spiral of silence accelerates the convergence of opinions and makes it easier to reach a consensus, but it also leads to more alienation of people and a decrease in the global clustering coefficient of the network. This study contributes to a deeper understanding of opinion dissemination in social networks and provides an important reference for opinion prediction.