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：12
页码：17528
DOI：10.15680/IJIRCCE.2017.0512042
出版社：S&S Publications
摘要：Today's market is the online market, most users prefer to do their own business via the Internet (such asonline shopping, etc.). Therefore, providing the best services for the user is the most difficult task. To address thisproblem, we focus on the peer-reviewed review model (user-generated review) and global qualificationsi.e.rating andtry to identify semantic aspects and aspect-level sentiments from data to review and anticipate the general sentiments ofreviews.We propose a probabilistic novel supervised joint aspect and sentiment model(SJASM) to treat problems at thesame time in a unitary framework. SJASM represents each review document in the form of pairs of opinions and cansimulate simultaneously the terms of appearance and the corresponding opinion words of the review for the hiddenaspect and thesentiment detection. It also uses global sentimental classifications, which often come online, such as datamonitoring, and can infer semantic aspects and feelings in terms of appearance that are not only meaningful, but evenpredictive of general feelings of revision. We have also designed a recommendation system, mainly a recommendationsystem that generates a cold start problem. Our system solves this problem through the use of collaboration techniques.
