Title :
Sentiment text classification of customers reviews on the Web based on SVM
Author :
Xia, Huosong ; Tao, Min ; Wang, Yi
Author_Institution :
Dept. of Econ. & Manage., Wuhan Univ. of Sci. & Eng., Wuhan, China
Abstract :
As a developing endeavor of data mining on semi-structured information, sentiment analysis to the comments on the Internet has aroused people´s great interest recently. This paper analysis the influence of different stop word removal methods on the result of text classification and represent the more effective stop word removal list. The experiment bases on the sentiment comments which have been grasped on the Web, using two different kinds of feature selection, choose the TF-IDF function to calculate the feature weights. Implement the classification with the technology of SVM.
Keywords :
Internet; classification; data mining; support vector machines; text analysis; Internet; TF-IDF function; World Wide Web; customer review; data mining; feature selection; semistructured information; sentiment analysis; sentiment comment; sentiment text classification; stop word removal method; support vector machine; Accuracy; Feature extraction; Internet; Support vector machines; Text categorization; Training; Vocabulary; feature selection; sentiment classification; stop words removal; support vector machine;
Conference_Titel :
Natural Computation (ICNC), 2010 Sixth International Conference on
Conference_Location :
Yantai, Shandong
Print_ISBN :
978-1-4244-5958-2
DOI :
10.1109/ICNC.2010.5584077
