Font Size: a A A

Research On The Emotional Portraits Of Weibo Users In Public Opinion Events

Posted on:2023-02-07Degree:MasterType:Thesis
Country:ChinaCandidate:Y W WangFull Text:PDF
GTID:2557307103479374Subject:Books intelligence
Abstract/Summary:PDF Full Text Request
In recent years,the popularity of social media has rapidly increased,and the number of social media users has surged.As one of the typical representatives of social media platforms,Weibo contains a large amount of user comment data,which provides data support for user sentiment analysis.When emergencies occur,Weibo users share their views and express their emotions through the platform.Among them,the rapid spread of negative emotions and extreme emotions will lead to the outbreak of public opinion,which not only destroys the harmony of cyberspace,but even endangers social security and stability.This paper introduces the background and significance of the research on Weibo users’ emotions in public opinion events,and sorts out the relevant research results of domestic and foreign scholars in detail,thus leading to the research theme of this paper,that is,the research on Weibo users’ emotional portraits in public opinion events.First of all,taking "Sun Wenbin’s killing of a doctor" as an example,we collect user comment texts and user-related information on Weibo platform,and after data processing is completed,based on the existing emotional dictionary,we build and improve the emotional dictionary in this field.Calculate the sentiment value of Weibo comments,initially identify the basic emotional tendencies of Weibo users,and use it as the user’s explicit label.Secondly,using the user portrait construction method of Ontology,construct emotional portraits of Weibo users in the event.Among them,the explicit tags set nicknames,numbers,comment time,text,and basic emotional tendencies,and the invisible tags set satisfaction and happiness,fear and panic,sadness and sympathy,anxiety and anger,stability and calm.reveal the user’s detailed emotional profile.Then,the emotional portrait characteristics of users are analyzed based on the three perspectives of public opinion development cycle,public opinion user subject and user clustering.The study concludes that in public opinion events,the hidden emotional value of users has different emphasis in different stages.The satisfaction value is low in the outbreak period of public opinion events and reaches the highest in the recession period.Users’ hidden feelings of fear,panic,sadness,sympathy,anxiety and anger are the most intense in the climax period of public opinion.At the same time,the emotional value of opinion leaders will be significantly higher than that of ordinary users,which is easy to incite more negative emotions and thus spread them to ordinary users.Based on this,this paper proposes emotional guidance strategies from three aspects: giving play to the guiding function of emotional portraits,implementing full-period emotional guidance and establishing a multi-subject cooperation mechanism,providing reference for netizens’ emotional guidance and the government’s public opinion governance.
Keywords/Search Tags:emotional portrait, sentiment analysis, public opinion, microblog users
PDF Full Text Request
Related items