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：3
页码：6376
DOI：10.15680/IJIRCCE.2017.0503414
出版社：S&S Publications
摘要：The main objective of this system is to analyze the mining techniques with user opinions regarding theirshopping reviews and recommendations to provide great support to both manufacturers and customers. Opinion Miningtargets and opinion words from online reviews for products, which are important tasks for fine grained opinion mining,the key component of which involves detecting opinion relations among words. To this end, this approach proposes anovel approach based on the partially supervised alignment model, which regards identifying opinion relations as analignment process. The distribution of polarity ratings over reviews written by different users or evaluated based ondifferent products for mining are often skewed in the mining industries. User reviews are highly essential for user andproduct information, which would be helpful for the task of sentiment classification of reviews. The temporal nature ofreviews posted by the same user or evaluated on the same product are ignored by the researchers in past. Finally, wefeed the user, product and review representations into a machine learning classifier for sentiment classification. Ourapproach has been evaluated on three large-scale review datasets from the IMDB and Yelp. In particular, compared tothe traditional unsupervised alignment model, the proposed model obtains better precision because of the usage ofpartial supervision. In addition, when estimating candidate confidence, we penalize higher-degree vertices in our graphbasedranking algorithm to decrease the probability of error generation. Our experimental results on three corpora withdifferent sizes and languages show that our approach effectively outperforms state-of-the-art methods.
