Content area
Full Text
World Wide Web (2015) 18:949967 DOI 10.1007/s11280-014-0289-x
A word-emoticon mutual reinforcement ranking model for building sentiment lexicon from massive collection of microblogs
Shi Feng Kaisong Song Daling Wang Ge Yu
Received: 26 August 2013 / Revised: 21 December 2013 / Accepted: 6 March 2014 / Published online: 27 April 2014 Springer Science+Business Media New York 2014
Abstract Recently, more and more researchers have focused on the problem of analyzing peoples sentiments and opinions in social media. The sentiment lexicon plays a crucial role in most sentiment analysis applications. However, the existing thesaurus based lexicon building methods suffer from the coverage problems when faced with the new words and new meanings in social media. On the other hand, the previous learning based methods usually need intensive expert efforts for annotating training datasets or designing extraction patterns. In this paper, we observe that the graphical emoticons are good natural sentiment labels for the corresponding microblog posts and a word-emoticon mutual reinforcement ranking model is proposed to learn the sentiment lexicon from the massive collection of microblog data. We integrate the emoticons and candidate sentiment words in the microblogs to construct a two-layer graph, on which a random walk is run for extracting the top ranked words as a sentiment lexicon. Extensive experiments were conducted on a benchmark dataset with various topics. The results validate the effectiveness of the proposed methods in building sentiment lexicon from microblog data.
Keywords Sentiment analysis Opinion mining Lexicon building Microblog mining
S. Feng ([envelopeback]) K. Song D. Wang G. Yu
Institute of Computer Software and Theory, Northeastern University, No.3-11 Wenhua Road, Heping District, Shenyang, Chinae-mail: mailto:[email protected]
Web End [email protected]
K. Songe-mail: mailto:[email protected]
Web End [email protected]
D. Wange-mail: mailto:[email protected]
Web End [email protected]
G. Yue-mail: mailto:[email protected]
Web End [email protected]
950 World Wide Web (2015) 18:949967
1 Introduction
As the rise of Web 2.0 technologies, more and more people are willing to publish their attitudes and feelings in Web 2.0 based social media rather than just passively browse and accept information. Because of the rich user-generated information in social media, how to provide an efficient way to analyze users sentiments has received significant attention from both academic researchers and commercial companies [29]. The sentiment analysis in social media includes subjectivity and polarity classification, sentiment holder...