期刊名称：International Journal of Computer Science and Network
印刷版ISSN：2277-5420
出版年度：2016
卷号：5
期号：2
页码：239-245
出版社：IJCSN publisher
摘要：The objective of the paper is to demonstrate theviability of analyzing online data. It displays a frameworkwhich after effects pattern investigation that will be shown asresults with various segments introducing positive, negativeand neutral. It is challenging task to summarize opinionabout the products due to diversity and size. Mining onlineopinion mining is a difficult text classification task ofsentiment analysis. Multivariate content technique calledFeature Relation Network that considers semantic data,influencing the syntactic connections between n-gramfeatures. FRN empowers the consideration of heterogeneousn-gram features for improved opinion classification, byjoining syntactic data about n-gram relations. FRN selectsthe features in a more computationally effective way thannumerous multivariate and hybrid methods. Appropriatefeature selection and representation with sentiment analysis,accuracies using support vector mechanism sentimentanalysis; the task of text pre-processing is to be explored.
关键词：Sentiment Analysis; Text pre-processing; Feature;Relation Network (FRN); Support Vector Machine (SVM).
