期刊名称：International Journal of Advanced Research in Computer Engineering & Technology (IJARCET)
印刷版ISSN：2278-1323
出版年度：2015
卷号：4
期号：4
页码：1173-1178
出版社：Shri Pannalal Research Institute of Technolgy
摘要：Data mining is the process of analyzing interesting data from different perspective and summarizing into useful information. Sentiment analysis is an application of natural language processing, data mining and text mining to identify sentiments or mood of the public about particular topic or products or customer reviews. This paper proposes to improve the methods for feature selection for customers review and also detect the polarity of reviews using machine learning approach then consider score as evidence for overall reviews weighting. Objective of the research paper is to select the best features selection methods for the aspect level of the sentiment analysis. Bog of noun, bog of words, stop word removal , parts of speech, ABC algorithm are used for feature set selection. For classification K-NN, Na.ve Bayes, Support vector machine clustering algorithms are used for classification of sentiment weighted analysis.
关键词：Sentiment analysis; Feature selection; ; Artifiacal bee colony algorithm; Term weighting; Support ; Vector Machine.
