An unsupervised approach to rank product reviews

Author

Jianwei Wu ; Bing Xu ; Sheng Li

Author_Institution

MOE-MS Key Lab. of Natural Language Process. & Speech, Harbin Inst. of Technol., Harbin, China

Volume

3

fYear

2011

fDate

26-28 July 2011

Firstpage

1769

Lastpage

1772

Abstract

With the development of online shopping, more and more product reviews are acquired from online shopping sites, which vary a wide range in quality. In order to solve the problem of detecting low-quality reviews, we view the problem as a ranking task and a link analysis based ranking method is proposed. The proposed method requires no domain knowledge and no training data. Experiment results indicate that the proposed approach is effective in (1) showing comparable performance with the SVM (Support Vector Machines) regression method and (2) domain independent.

Keywords

Internet; Web sites; regression analysis; retail data processing; support vector machines; unsupervised learning; SVM regression method; link analysis based ranking method; online shopping site; rank product review; support vector machine; unsupervised approach; Algorithm design and analysis; Data mining; Digital audio players; Feature extraction; Natural language processing; Support vector machines; Training data; link analysis; opinion mining; review quality detection;

fLanguage

English

Publisher

ieee

Conference_Titel

Fuzzy Systems and Knowledge Discovery (FSKD), 2011 Eighth International Conference on

Conference_Location

Shanghai

Print_ISBN

978-1-61284-180-9

Type

conf

DOI

10.1109/FSKD.2011.6019793

Filename

6019793