Sentiment analysis on Weibo data

Author

Di Li ; Jianwei Niu ; Meikang Qiu ; Meiqin Liu

Author_Institution

State Key Lab. of Software Dev. Environ., Beihang Univ., Beijing, China

fYear

2014

fDate

20-22 Oct. 2014

Firstpage

249

Lastpage

254

Abstract

With the development of the Internet, people share their emotion statuses or attitudes on online social websites, leading to an explosive rise on the scale of data. Mining sentiment information behind these data helps people know about public opinions and social trends. In this paper a sentiment analysis algorithm adapting to Weibo (Microblog) data is proposed. Given that a Weibo post is usually short, LDA model is used to generate text features based on semantic information instead of text structure. To decide the sentiment polar and degree, SVR model is used here. Experiment shows the algorithm performs well on Weibo data.

Keywords

data mining; natural language processing; social networking (online); text analysis; Internet; LDA model; SVR; Weibo data; Weibo post; attitudes; emotion statuses; microblog; online social Web sites; public opinions; semantic information; sentiment analysis algorithm; sentiment degree; sentiment information mining; sentiment polar; social trends; text features; Algorithm design and analysis; Classification algorithms; Feature extraction; Hidden Markov models; Kernel; Sentiment analysis; Support vector machines; public opinion monitoring; sentiment analysis; text classification;

fLanguage

English

Publisher

ieee

Conference_Titel

Computing, Communications and IT Applications Conference (ComComAp), 2014 IEEE

Conference_Location

Beijing

Print_ISBN

978-1-4799-4813-0

Type

conf

DOI

10.1109/ComComAp.2014.7017205

Filename

7017205