Evolution of public opinion based on network structure balance

With the development of internet technology, social network has become an important platform for people to get information and communicate with each other. The current researches focus on the identification of opinion leaders in the public opinion network and the analysis of viewpoints between individuals with bounded confidence models. However, few scholars study the evolution of public opinion through theory of network structure balance. Therefore, we propose a novel method based on network structure balance to analyze the evolution of public opinion. First, we establish a diversity evaluation method to calculate node's influence by the comment, forward and praise. Second, we construct the network with positive and negative weights by node's attitude to hot event and the interaction forces among nodes. At last, we merge the balance triangles with the same positive and negative attribute as new nodes, and explain the evolution of public opinion on the network. The analysis of public opinion based on structural balance provides a new way to understand the evolution of public opinion.