期刊名称：International Journal of Innovative Research in Computer and Communication Engineering
印刷版ISSN：2320-9798
电子版ISSN：2320-9801
出版年度：2017
卷号：5
期号：10
页码：15943
DOI：10.15680/IJIRCCE.2017.0510050
出版社：S&S Publications
摘要：Due to the huge development of internet and social network, the data sources are widely increased.Number of reviews and feedbacks for a particular product or service has been shared by the users via social networks.Finding the best product or service from learning the reviews is more tedious to the users. Data mining is the effectiveway to handle such huge data reviews and allows the user to get the best product according to the reviews. There areseveral solutions in the existing system performed opinion and sentiment mining from the social data’s, which poseshuge set of drawbacks and complications. Finding the product or service rating from the review analysis involved withtextual analysis. This paper developed a new sentiment based rating predication approach for mobile applications fromhuge number of reviews. Predicting, ranking and rating the mobile apps and its services with the considerations ofsemantic and hyponym features. This gives a semantic expression to improve the classification accuracy and theproposed technique was investigated on four classifiers. There was a uniform improvement in the classificationaccuracy for all the classifiers tested. The output of the application is experimented and tested with the mobile appreview dataset crawled form Google play store. The use of Hybrid sentiment analysis methods with combining aspect,sentence and document levels were performed in the proposed system.
