Social Network Public Opinion Research Based on S-SEIR Epidemic Model

Abstract

This paper proposes an S-SEIR-based social network public opinion prediction model. The model takes into account the derivative nature of information in the micro-blog network, as well as the unique propagation rules and textual sentiments in micro-blog. The S-SEIR model of microblog information propagation based on user text sentiment constructed. The factors affecting information dissemination in microblog were studied. The simulation results show that the emotion of the text can directly affect the propagation behavior of the node. Besides, considering the emotional factors of the text can also significantly improve the accuracy of public opinion trend prediction.