Research On Internet Public Opinion Diffusion Model

In this era of informationization and dataization,the Internet has become another platform for people to gain knowledge and information.With the development of the network,a new expression of public opinion was born-the online public opinion.The degree of spread of online public opinion events on the Internet;the direction of public opinion spread;and people’s attitudes to public opinion events have an increasing impact on social economy and social stability.Therefore,the analysis and prediction of the spread and spread of online public opinion will help the government to manage and control the degree of public opinion development,correctly handle and guide public opinion,and reduce and avoid negative impacts.This article uses literature research methods,case analysis methods,and other research methods,using traditional infectious disease models as entry points,combined with statistical methods to analyze and study the spread and diffusion process and trends of online public opinion.Taking the spread of Internet public opinion spread on Sina Weibo as an example,the full text establishes two different prediction models of spread of online public opinion spread.One is the traditional model of spread of public opinion spread and uses parameter inversion to estimate the parameters of the traditional public opinion spread model Use the estimated parameters to make predictions.One is the gray prediction model and uses the support vector machine to correct the prediction residuals obtained from the gray prediction,and compares and analyzes the two network public opinion models.Finally,this article takes the case of the Liangshan Mountain Fire on Sina Weibo and the case of the new type of coronary pneumonia as examples,and performs data crawling and model analysis using MATLAB software.Empirical analysis of theoretical models.According to the selected cases,both the traditional infectious disease model and the gray support vector machine model can predict the direction and extent of the public opinion diffusion that is about to end.However,because the traditional infectious disease model requires a large amount of data support,it is impossible to accurately predict the ongoing public opinion.The gray support vector machine model can accurately predict whether to end public opinion.