Home // ACHI 2015, The Eighth International Conference on Advances in Computer-Human Interactions // View article

Sentiment Classification for Chinese Microblog

Authors:
Wen-Hsing Lai
Chang-Hsun Li

Keywords: sentiment classification; emotion; support vector machine; microblog; emoticon

Abstract:
A sentiment classification method for Chinese microblog is presented. For short sentence microblog, it is very challenging because the information of emotion is very limited. First, an emotion lexicon is built from training corpus. A simple measure – difference ratio is used to choose words from lexicon as features for classification. Support vector machine and voting on counts and accumulated difference ratio are jointly combined as classification method. The experimental results show that our recognition rate is better than the popular method using collocation strength. Our recognition improvement is about 2.06% in testing. Therefore, the difference ratio measure we used and the tactic in constructing the lexicon are proved very effective.

Pages: 219 to 223

Copyright: Copyright (c) IARIA, 2015

Publication date: February 22, 2015

Published in: conference

ISSN: 2308-4138

ISBN: 978-1-61208-382-7

Location: Lisbon, Portugal

Dates: from February 22, 2015 to February 27, 2015