Study On The Emotional Evolution Of Network Public Opinion In Major Public Health Emergencies

Major public health emergencies are related to the vital interests of every citizen,usually,their harmful consequences will quickly spread to online media platforms,causing serious public opinion crises in real life and the online world.Therefore,relying on the "hot" public opinion information to analyze the sentiment of netizens,provide corresponding suggestions and support for the government to quickly supervise online public opinion and grasp social pain points.In this paper,COVID-19 as an example,sentiment analysis of network public opinion information.The main research work is as follows.Firstly,use self-taught Python knowledge to write a web crawler program to obtain the microblog and comments related to the epidemic released by CCTV news,and to sort and analyze the obtained data.Secondly,based on the emotional dictionary of Harbin Institute of Technology,the emotional dictionary was expanded by using the COVID-19 Corpus to obtain an emotional dictionary suitable for this study.Train sentiment tendency analysis model and test the model effect.Thirdly,we classify the obtained microblog texts and explore the evolution of microblogs with different themes every day.The sentiment types of the obtained comment text data are judged and the word frequency of sentiment words is counted,and the user sentiment is mined and visualized.Fourth,combined the development of the "COVID-19" divided into evolutionary stages,the sentiment evolution characteristics and distribution of each stage were analyzed.In this paper,After analyzing the characteristics and laws of emotional evolution at each stage,this thesis obtains the following laws:(1)The number of microblogs and their comments in the initial response period are relatively small.Internet users do not know enough about COVID-19,and the attitudes towards events are complex.There are two sentiment orientations,namely,blind self-confidence and excessive fear;(2)During the rapid growth period,the number of Weibo posts increased sharply,user participation was the highest,netizens had a deep understanding of the epidemic,and the proportion of fear emotions increased.Relevant departments should pay more attention to the evolution of Internet public sentiment during the rapid growth period,and timely popularize science to eliminate public misunderstandings;(3)During the high-frequency release period,the number of microblogs released by CCTV increased,and the degree of publicity of epidemic prevention information was high.At this stage,netizens mostly expressed derogatory accusations against people and things that were not conducive to prevention and control.Relevant departments should monitor the derogatory accusations in public opinions in a timely manner,and put an end to things that were not conducive to epidemic prevention and control as early as possible.(4)During the period of fluctuation decline and slow recovery,the epidemic prevention and control work has become normalized.The relevant departments need to timely and properly solve the adverse events concerned in public opinion and publicly dispose the results to avoid further expansion of public opinion events.