摘要：

Abstract:

【Purpose/significance】Since the We-media era, reversal of public opinion events has occurred frequently. Studying the in⁃

fluencing factors and predicting the possibility of public opinion reversal have important practical significance for early detection and

effective avoidance of the risk of public opinion reversal.【Method /process】By analyzing the influencing factors of public opinion re⁃

versal, based on SVM, a prediction model of reversal We-media public opinion is established, and with python 3.0 to verify 33

We-media public opinion events.【Result/conclusion】The result shows that the model has accuracy and validity, and can better pre⁃

dict the possibility of public opinion reversal.【Innovation/limitation】Based on previous opinions, seven factors including the nature

of public opinion events, the popularity of public opinion, the authority of the first subject of the public opinion, the form of public

opinion dissemination, and the emotional tendency of netizens were put forward, and the forecast model was constructed through calcu⁃

lation of indicators. In the later stage, the accuracy of the model can be improved by increasing the influencing factors of public opin⁃

ion reversal and adjusting model parameters.