关键词: 网络舆情治理, 大规模群体决策, SEIR演化模型

Abstract: In the context of the new media era, the Internet has become an indispensable presence in people’s lives, and it is very easy for unexpected crisis events to ferment rapidly in cyberspace, forming a cyber public opinion crisis, which in turn threatens social stability. When an online public opinion crisis occurs, if the government fails to effectively manage online public opinion and stop the conflict of public opinion, it will cause public order chaos and social panic. The development of online public opinion is jointly influenced by multiple participants such as official media, internet celebrity and netizens, and due to the convenience of social networking platforms, the online public opinion spreads very rapidly, which is undoubtedly a challenge for the government. Therefore, in the context of promoting the modernization of the national governance system and governance capacity, in-depth research is needed on how to quickly and effectively manage online public opinion.
Based on the characteristics of multi-point outbreaks of sudden crisis events and the promotional effect of internet celebrity on the evolution of events, this paper adopts the large-scale group decision-making and SEIR evolution models to study the formation mechanism and evolution process of network public opinion respectively, analyzes the threshold of public opinion dissemination and the evolution trend in different conditions, and studies the selection of strategies for network public opinion management under different network public opinion heat levels. The selection of network public opinion management strategy is studied, and the effectiveness and scientificity of the method are verified through case simulation. In this paper, we use python to capture the popular comments of microblogging topics to obtain real data processing and calculation to get the Internet celebrity’s behavior trend. After that, we simulate the actual development process of emergencies by using the SEIR model and analyze the impact of different behavioral decisions of the two on the emotional state of netizens, as well as the mechanism of network public opinion formation and evolution process. In addition, effective strategies are provided for the prevention and management of online public opinion crisis.
This study investigates the online public opinion governance of sudden crisis events based on large-scale group decision-making and SEIR evolution model, and explores the evolution of netizens’ emotional state under different heat levels. Firstly, at an early stage of a sudden public crisis event, when the heat of public opinion is low, it is crucial for the official media to detect and guide the public opinion in a timely manner, so as to prevent the public opinion from spreading and to manage the event as early as possible, thus saving time and resources. Secondly, when the online public opinion is hot to a certain extent, the governance is more difficult, so the government needs to increase the authority of official reports and control the speech norms of internet celebrity, so as to realize the effective governance of the online public opinion. Finally, it is necessary to strengthen netizens’ education on the quality of the Internet and improve their ability to distinguish right from wrong, so that they can correctly identify the information in the event of Internet public opinion and reduce the possibility of the vicious evolution of Internet public opinion.
The significance of further research is that monitoring the development of online public opinion helps online public opinion governance. For example, the heat level of online public opinion dissemination is monitored to capture changes in the heat level. In this way, appropriate means of handling can be selected at different heat levels, which can better stop the vicious development of online public opinion, while reducing the investment of material and human resources and saving resources. In addition, better monitoring of online public opinion can detect crises in a timely manner, giving the relevant departments enough time to react and manage them better.

Key words: online public opinion governance, large-scale group decision making, SEIR evolutionary modeling