Publisher： IGI Global_journal
E-ISSN： 1947-8216|8|3|34-44
ISSN： 1947-8208
Source： International Journal of Knowledge and Systems Science (IJKSS), Vol.8, Iss.3, 2017-07, pp. : 34-44
Disclaimer: Any content in publications that violate the sovereignty, the constitution or regulations of the PRC is not accepted or approved by CNPIEC.
Abstract
In this paper, the authors investigate the topic-independent Chinese sentiment identification problem from online news. They analyze the word usage and sentence structure of the documents for inferring representative terms and sentences in the documents, and then employ the feature values of each document for identifying the opinion of the topic-independent online news. The support vector machine (SVM) is leveraged for training the classified model in terms of the extracted features and identifying the opinion orientation of the topic-independent documents by the trained model. Experimental results demonstrated that the authors' features are helpful for identifying the opinions of the topic-independent documents, and can help readers for filtering out the negative documents.
Related content
Access to resources
Recommend
