DocumentCode :
1442211
Title :
Aspect-Based Opinion Polling from Customer Reviews
Author :
Zhu, Jingbo ; Wang, Huizhen ; Zhu, Muhua ; Tsou, Benjamin K. ; Ma, Matthew
Author_Institution :
Key Lab. of Med. Image Comput. (Minist. of Educ.), Northeastern Univ., Shenyang, China
Volume :
2
Issue :
1
fYear :
2011
Firstpage :
37
Lastpage :
49
Abstract :
Opinion polling has been traditionally done via customer satisfaction studies in which questions are carefully designed to gather customer opinions about target products or services. This paper studies aspect-based opinion polling from unlabeled free-form textual customer reviews without requiring customers to answer any questions. First, a multi-aspect bootstrapping method is proposed to learn aspect-related terms of each aspect that are used for aspect identification. Second, an aspect-based segmentation model is proposed to segment a multi-aspect sentence into multiple single-aspect units as basic units for opinion polling. Finally, an aspect-based opinion polling algorithm is presented in detail. Experiments on real Chinese restaurant reviews demonstrated that our approach can achieve 75.5 percent accuracy in aspect-based opinion polling tasks. The proposed opinion polling method does not require labeled training data. It is thus easy to implement and can be applicable to other languages (e.g., English) or other domains such as product or movie reviews.
Keywords :
customer satisfaction; data mining; natural language processing; text analysis; Chinese restaurant reviews; aspect identification; aspect-based opinion polling; aspect-based segmentation model; customer reviews; customer satisfaction; multiaspect bootstrapping method; multiaspect sentence segmentation; opinion mining; unlabeled free-form textual customer reviews; Classification algorithms; Customer service; Subspace constraints; Training data; Opinion polling; aspect-based analysis.; opinion mining; sentiment analysis;
fLanguage :
English
Journal_Title :
Affective Computing, IEEE Transactions on
Publisher :
ieee
ISSN :
1949-3045
Type :
jour
DOI :
10.1109/T-AFFC.2011.2
Filename :
5708129
Link To Document :
بازگشت