面向舆情大数据的网民情绪演化机理及趋势预测研究

Research on the Evolution Mechanism and Trend Prediction of Netizens for Public Opinion Big Data

[目的/意义]大数据环境下,不同的网民情绪在舆情传播过程中相互影响,有些甚至产生"群体极化"现象,如何准确把握网民情绪演化机理并预测其演化趋势是大数据环境下网络舆情治理的重要内容。[方法/过程]定性分析大数据环境下网民情绪特征和分类,构建网民情绪演化机理的微分方程模型,通过计算模型平衡点及其稳定性条件研究网民情绪竞争场景以及通过回归分析研究网民情绪演化趋势预测问题。[结论/结果]经过理论建模和仿真分析得出不同类型情绪网民数量的变化规律,得出面向网络舆情的网民情绪趋势预测方法,最后提出网民情绪演化机理模型高维拓展问题、同时期舆情事件对网民情绪的冲击问题以及网络舆情信息内容和网络推手等对网民情绪的影响问题等一系列有待进一步研究的问题。

[Purpose/Significance]In big data network public opinion environment, different users＇ emotions interact in the process of public opinion dissemination, some even show a ＂ group polarization＂ phenomenon, and how to accurately grasp the evolution mechanism for netizens＇ emotions and predict the evolution trend are the important content of network public opinion management under the big data environment. [ Method/Process] Qualitative analysis is used to explore the netizens＇ emotional characteristics and classification under the big data environment, the differential equation model for emotion evolution mechanism of netizens is constructed, the netizen emotions competitive scene is studied via calculating the model balance point and discussing its stability conditions, and the prediction of the evolu-tion trend of netizens＇ emotions is analyzed through regression analysis. [ Result^Conclusion] Through theoretical modeling and simulation analysis, the change rules for the numbers of the netizens of different types of emotions are identified, and the emotions trend prediction methods oriented to Internet public opinion are summarized. The research also poses the issues for further study such as the netizen emo-tions evolution mechanism model of higher dimensional expansion, the impact of other same period public opinion events on the netizens 爷 emotions, and the influence of network public opinion information content and Internet marketer on netizens＇ emotions.