Learning sentiment classification model from labeled features

Research output: Chapter in Book/Report/Conference proceedingOther chapter contribution

View graph of relations Save citation

Authors

Research units

Abstract

We propose a novel framework where an initial classifier is learned by incorporating prior information extracted from an existing sentiment lexicon. Preferences on expectations of sentiment labels of those lexicon words are expressed using generalized expectation criteria. Documents classified with high confidence are then used as pseudo-labeled examples for automatical domain-specific feature acquisition. The word-class distributions of such self-learned features are estimated from the pseudo-labeled examples and are used to train another classifier by constraining the model's predictions on unlabeled instances. Experiments on both the movie review data and the multi-domain sentiment dataset show that our approach attains comparable or better performance than exiting weakly-supervised sentiment classification methods despite using no labeled documents.

Request a copy

Request a copy

Details

Publication date2010
Publication titleProceeding : CIKM '10 proceedings of the 19th ACM international conference on information and knowledge management
Place of PublicationNew York (US)
PublisherACM
Pages1685-1688
Number of pages4
ISBN (Print)978-1-4503-0099-5
Original languageEnglish
Event19th ACM international conference on information and knowledge management, CIKM '10 - Toronto, Canada

Conference

Conference19th ACM international conference on information and knowledge management, CIKM '10
CountryCanada
CityToronto
Period26/10/1030/10/10

    Keywords

  • sentiment analysis, opinion mining, generalized expectation, self-learned features, weakly-supervised classification

DOI

Employable Graduates; Exploitable Research

Copy the text from this field...