Font Size: a A A

Risk Assessment Model Of Weibo Online Public Opinion Based On Sentiment Analysis

Posted on:2021-02-21Degree:MasterType:Thesis
Country:ChinaCandidate:L GengFull Text:PDF
GTID:2428330629988924Subject:Engineering
Abstract/Summary:PDF Full Text Request
The rapid growth of the mobile Internet has fundamentally changed people's life.There is an increase in people using the Internet to publish personal opinions and share their private life.As a mature social networking platform with hundreds of millions of internet users,Weibo generates data covering all ages,industries,and levels.It plays a pivotal role in the development of China's online public opinion and has enormous potential commercial value and social value.Weibo short texts has the characteristics of fragmentation and easy dissemination,which provides a hotbed for the generation and spread of Internet rumors.A healthy development of online public opinion will also play a positive role in social stability.Therefore,taking sentiment analysis and public opinion monitoring of Weibo text is becoming a top priority.The purpose of this paper is to conduct a deep public opinion analysis of the Weibo short texts,that is,quickly extract key information from the disordered Weibo short texts and analyze their affective tendencies.The content of this paper is divided into the following aspects:(1)Construction of sentiment dictionary on Weibo.Weibo short text belongs to the web text but has the characteristics of being unique to the microblog.By processing the original Weibo text and selecting emotional words with the characteristics of Weibo,the domain-specific sentiment dictionary combining the basic sentiment dictionary and the network sentiment dictionary,which enables sentiment analysis more precisely.(2)Construction of Weibo public opinion monitoring model.Based on the emotion classification model of the sentiment lexicon of Weibo,the online public opinion events are monitored.The comprehensive emotional score of the topic is calculate according to algorithm the sentiment tendency of the sentence and the sentiment intensity of the topics.Adding time stamps and obtains the sentiment tendency and sentiment change trend in different time stamps.By understanding the topic of the Weibo emotional evolution process,this article introduces the public opinion rating of Weibo topics through five dimensions.(3)Empirical analysis of Weibo.The model analyzed the examples of Weibo public opinion.The way to evaluate the model through evaluation indicators such as Accuracy,Recall,and F-measure,and verify the validity of the model in conjunction with Weibo samples.
Keywords/Search Tags:Sentiment analysis, Sentiment lexicon, Emotional intensity, Online public opinion, Short text
PDF Full Text Request
Related items