The following information was submitted:
Transactions: WSEAS TRANSACTIONS ON INFORMATION SCIENCE AND APPLICATIONS
Transactions ID Number: 32-249
Full Name: Ziqiong Zhang
Position: Ph.D. Candidate
Age: ON
Sex: Female
Address: School of Hotel & Tourism Management, Hong Kong Polytechnic University, HONG KONG
Country: HONG KONG S.A.R.
Tel: 00852-51358283
Tel prefix:
Fax:
E-mail address: hmzhang@inet.polyu.edu.hk
Other E-mails: xiaojia0459@yahoo.com.cn
Title of the Paper: Sentiment Classification to Online Noisy Cantonese Reviews
Authors as they appear in the Paper: Ziqiong Zhang, Qiang Ye, Rob Law, Yijun Li
Email addresses of all the authors: {hmzhang@polyu.edu.hk, hmye@polyu.edu.hk, hmroblaw@polyu.edu.hk, liyijun@hit.edu.cn
Number of paper pages: 11
Abstract: Cantonese is an important Chinese dialect spoken in some regions of Southern China. Local online users often represent their opinions and experiences with written Cantonese on the Web. The information in these reviews is valuable to both manufacturers and potential consumers. Although search engines like Google can help people find most of reviews about certain product, it is difficult, if not impossible, to personally read all the returned pages by any one. Sentiment classification would be helpful by automatically classifying reviews of customers for a product or service as recommended (thumbs up) or not recommended (thumbs down). In this research, sentiment classification techniques were incorporated into the very noisy domain of online Cantonese-written restaurant reviews. Using two machine learning-based classifiers, this paper conducted a series of experiments to explore appropriate methods for automatic sentiment classification in the very noisy domain of on!
line Cantonese-written reviews. The results show that, the support vector machine classifier based on a Mandarin Chinese word segmentation tool performed surprisingly well. The accuracy, precision and recall respectively for positive and negative reviews all reach above 85%, when the training corpus contained 5000 or more reviews.
Keywords: Text mining, Sentiment classification, Online reviews, Cantonese, Restaurants
EXTENSION of the file: .doc
Special (Invited) Session:
Organizer of the Session:
How Did you learn about congress:
IP ADDRESS: 158.132.12.80
