PUBLIC OPINION FORMATION MODEL BASED ON OPINION ENTROPY

Abstract

In this paper, we introduce the concept of opinion entropy based on Shannon entropy, which is used to describe the uncertainty of opinions. With opinion entropy, we further present a public opinion formation model, and simulate the process of public opinion formation under various controlled conditions. Simulation results on the Holme–Kim network show that the opinion entropy will reduce to zero, and all individuals will hold the opinion of agreeing with the topic, only by adjusting the cons' opinions with a high control intensity. Controlling the individuals with big degree can bring down the opinion entropy in a short time. Besides, extremists do not easily change their opinion entropy. Compared with previous opinion clusters, opinion entropy provides a quantitative measurement for the uncertainty of opinions. Moreover, the model can be helpful for understanding the dynamics of opinion entropy, and controlling the public opinion.