By Topic

Sentiment Classification for Chinese Netnews Comments Based on Multiple Classifiers Integration

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

3 Author(s)
Wen Fan ; Sch. of Comput. Sci., Commun. Univ. of China, Beijing, China ; Shutao Sun ; Guohui Song

With the development of World Wide Web technologies, more and more netizens express their opinions on society and politics in net news comments. Sentiment classification is one of the most important sub-problems of opinion mining, which can classify net news comments as positive or negative to help government automatically identify the netizens' viewpoints on news event and make right decision or help enterprises find out weather the customers satisfy the products or not. Most of the researches for sentiment classification only use single classifier, such as kNN, Naive Bayes and Support Vector Machine (SVM). In this paper, we use two multiple classifiers integration algorithms, which are Bagging and Boosting, to conduct the sentiment classification. Different feature selection methods are also investigated. The result of experiment shows that AdaBoost approach, a type of Boosting, usually achieve better performance than Bagging and single classifier and feature selection based on statistic is better than POS-based method for sentiment classification of Chinese net news comments.

Published in:

Computational Sciences and Optimization (CSO), 2011 Fourth International Joint Conference on

Date of Conference:

15-19 April 2011