期刊名称：International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
印刷版ISSN：2278-1323
出版年度：2013
卷号：2
期号：6
页码：2165-2172
出版社：Shri Pannalal Research Institute of Technolgy
摘要：Sentiment analysis is the task of identifying whether the opinion expressed in a document is positive or negative about a given topic. In sentiment analysis, feature reduction is a strategy that aims at making classifiers more efficient and accurate. Unfortunately, the huge number of features found in online reviews makes many of the potential applications of sentiment analysis infeasible. In this paper we evaluate the effect of a feature reduction method with both Support Vector Machine and Naive Bayes classifiers. The feature reduction method used is principle component analysis. Our results show that it is possible to maintain a state-of-the art classification accuracy while using less number of the features through Receiver operating characteristic curves and accuracy measures.
关键词：sentiment; opinion; feature; learning; support vector
