The Open UniversitySkip to content

Latent sentiment model for weakly-supervised cross-lingual sentiment classification

He, Yulan (2011). Latent sentiment model for weakly-supervised cross-lingual sentiment classification. In: 33RD European Conference on Information Retrieval (ECIR 2011), 18 - 21 Apr 2011, Dublin, Ireland.

Google Scholar: Look up in Google Scholar


In this paper, we present a novel weakly-supervised method for crosslingual sentiment analysis. In specific, we propose a latent sentiment model (LSM) based on latent Dirichlet allocation where sentiment labels are considered as topics. Prior information extracted from English sentiment lexicons through machine translation are incorporated into LSM model learning, where preferences on expectations of sentiment labels of those lexicon words are expressed using generalized expectation criteria. An efficient parameter estimation procedure using variational Bayes is presented. Experimental results on the Chinese product reviews show that the weakly-supervised LSM model performs comparably to supervised classifiers such as Support vector Machines with an average of 81% accuracy achieved over a total of 5484 review documents. Moreover, starting with a generic sentiment lexicon, the LSM model is able to extract highly domainspecific polarity words from text.

Item Type: Conference Item
Copyright Holders: 2011 The Author
Keywords: Latent sentiment model (LSM); cross-lingual sentiment analysis; generalized expectation; latent Dirichlet allocation
Academic Unit/Department: Knowledge Media Institute
Interdisciplinary Research Centre: Centre for Research in Computing (CRC)
Item ID: 28543
Depositing User: Kay Dave
Date Deposited: 18 Apr 2011 08:18
Last Modified: 28 Mar 2016 15:51
Share this page:

▼ Automated document suggestions from open access sources

Actions (login may be required)

Policies | Disclaimer

© The Open University   + 44 (0)870 333 4340