Paper: A Review Selection Approach for Accurate Feature Rating Estimation

ACL ID C10-2088
Title A Review Selection Approach for Accurate Feature Rating Estimation
Venue International Conference on Computational Linguistics
Session Poster Session
Year 2010
Authors

In this paper, we propose a review se- lection approach towards accurate esti- mation of feature ratings for services on participatory websites where users write textual reviews for these services. Our approach selects reviews that compre- hensively talk about a feature of a service by using information distance of the re- views on the feature. The rating estima- tion of the feature for these selected re- views using machine learning techniques provides more accurate results than that for other reviews. The average of these estimated feature ratings also better rep- resents an accurate overall rating for the feature of the service, which provides useful feedback for other users to choose their satisfactory services.