Extracting Domain-Dependent Semantic Orientations of Latent Variables for Sentiment Classification
SCIE
SCOPUS
- Title
- Extracting Domain-Dependent Semantic Orientations of Latent Variables for Sentiment Classification
- Authors
- Lee Y; Kim J; Lee J.-H.
- Date Issued
- 2009-03
- Publisher
- Springer
- Abstract
- Sentiment analysis of weblogs is a challenging problem. Most previous work utilized semantic orientations of words or phrases to classify sentiments of weblogs. The problem with this approach is that semantic orientations of words or phrases are investigated without considering the domain of weblogs. Weblogs contain the author's various opinions about multifaceted topics. Therefore, we have to treat a semantic orientation domain-dependently. In this paper, we present an unsupervised learning model based on aspect model to classify sentiments of weblogs. Our model utilizes domain-dependent semantic orientations of latent variables instead of words or phrases, and uses them to classify sentiments of weblogs. Experiments on several domains confirm that our model assigns domain-dependent semantic orientations to latent variables correctly, and classifies sentiments of weblogs effectively.
- URI
- https://oasis.postech.ac.kr/handle/2014.oak/35951
- DOI
- 10.1007/978-3-642-00831-3_19
- ISSN
- 0302-9743
- Article Type
- Article
- Citation
- LECTURE NOTES IN COMPUTER SCIENCE, vol. 5459/2009, page. 201 - 212, 2009-03
- Files in This Item:
- There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.