Online Public Opinion Deduction Based on an Innovative Cellular Automata

Abstract

As the Internet entered the 2.0 era, online users have become the main part of public opinion, and significant events happened around the world quickly disseminate through the Internet. Some negative comments usually are mixed into posts and articles. If let public opinion develop at will, some serious social events may trigger very fierce online volatility and they will have adverse effects on social stability. Therefore, to simulate the propagation mode of topic events on the Internet, and to predict the trending of public opinion on the entire Internet, this paper constructs a new type of model based on the idea of cellular automata and applies it to the online public opinion situation deduction. We arrange the nodes in new ways. An individual user is represented by a cell node in our model, and we build the cellular space based on directed graph to represent social network topology that users follow about each other in an innovative way. We delineate the users’ portrait based on the users’ attributes, design evolutionary rules based on the behaviors of users on the Internet and the law of information dissemination. Experiments with real data of online users showed that the proposed model in this paper had excellent performance, it not only reflected the law of information dissemination on the Internet, but also its deduction result was similar to the trending of real historical events.