Font Size: a A A

Sentiment Analysis Of LGBT-related New Media Events

Posted on:2019-02-13Degree:MasterType:Thesis
Country:ChinaCandidate:P F HuFull Text:PDF
GTID:2417330542984736Subject:Journalism
Abstract/Summary:PDF Full Text Request
The Internet,particularly microblogs,has been used to raise issues in the public sphere.Since there is low visibility for LGBT people in the traditional media in China,the Internet and new media platforms has played an active role to produce objective portrayals of LGBT people and remove prejudice.Regardless of their sexual orientation and/or gender identity,more and more people are offering outspoken support for the rights of LGBT people either through online or offline channels.Public incidents related to LGBT community can sometimes arouse heated discussion on social media platforms.An analysis into LGBT-related new media events on Sina Weibo,may give insights into why particular events resonate with the population in a context of a few political opportunities.Value-added theory,proposed by Neil Smelser,suggests that the generalized belief played a central role in the development of collective behavior.It is the socio-psychological factor that account for the occurrence of episodes of collective behavior.Smelser categorized generalized beliefs into five forms:hysterical belief,wish-fulfillment belief,hostile belief,norm-oriented belief and value-oriented belief.Norm and value-oriented beliefs displays more purposive collective efforts to redefine and restructure social environment.In this study,I examined and compared sentiments accumulated on Weibo during the outbursts of two types of LGBT-related new media events represented by norm and value-oriented beliefs respectively.I built a LGBT-related Weibo events database including 22 important Weibo events from 2014 to 2017.Two independent coders coded the types of leading generalized beliefs in those events.I then collected data including trending posts and comments below using keywords of the event with Python and cleaned the data.The data for further analysis comprised 67415 Weibo posts and comments.I used LIWC,a software widely applied in sentiment analysis,to calculate the percentage of affect words in each event.The results revealed that LGBT-related new media events led by value-oriented beliefs were more likely to elicit overall sentiments.The results also suggested that Weibo users displayed more positive feelings when the LGBT-related events occurred in the name of restructuring values than norms.Findings from this study implied on how the value-added theory,proposed 50 years ago,can be invoked to explain collective behavior based on the Internet in China.
Keywords/Search Tags:New Media Event, Sentiment Analysis, LGBT
PDF Full Text Request
Related items