The rapid development of social media services has facilitated the communication of opinions through microblogs/tweets, instantmessages, online news, and so forth. This article concentrates on the mining of emotions evoked by short text materials. Compared to the classical sentiment analysis from long text, sentiment analysis of short text is sometimes more meaningful in social media. We propose an intensive maximum entropy model for sentiment classification, which generates the probability of sentiments conditioned to short text by employing intensive feature functions. Experimental evaluations using real-world data validate the effectiveness of the proposed model on sentiment classification of short text. Copyright © 2015 Springer International Publishing Switzerland.
|Title of host publication||Database systems for advanced applications: DASFAA 2015 International Workshops, SeCoP, BDMS, and posters, Hanoi, Vietnam, April 20-23, 2015, revised selected papers|
|Editors||An LIU, Yoshiharu ISHIKAWA, Tieyun QIAN, Sarana NUTANONG, Muhammad Aamir CHEEMA|
|Place of Publication||Cham|
|Publication status||Published - 2015|
CitationRao, Y., Li, J., Xiang, X., & Xie, H. (2015). Intensive maximum entropy model for sentiment classification of short text. In A. Liu, Y. Ishikawa, T. Qian, S. Nutanong, & M. A. Cheema (Eds.), Database systems for advanced applications: DASFAA 2015 International Workshops, SeCoP, BDMS, and posters, Hanoi, Vietnam, April 20-23, 2015, revised selected papers (pp. 42-51). Cham: Springer.
- Sentiment classification
- Short text analysis
- Intensive maximum entropy model