文章基本信息

标题：Prediction of Online Products by Hybrid Feedback Based Recommendation System
本地全文：下载
作者：S.Aasha ; Prof. Dr. A.R. Mohamed Shanavas
期刊名称：International Journal of Innovative Research in Science, Engineering and Technology
印刷版ISSN：2347-6710
电子版ISSN：2319-8753
出版年度：2017
卷号：6
期号：8
页码：16209
DOI：10.15680/IJIRSET.2017.0608091
出版社：S&S Publications
摘要：Online shopping has been known for a rapidly growing business, although online mobile shopping hasnot followed these same growth patterns in the past, but it is now being recognized for its potential. Recommendersystems help the customers to find accurate product from a large database. For recommendation, admin train a databasewhich has sentiment based keywords with positivity or negativity weight, so that when the user post comments for aproduct, these sentimental keywords are mined from user’s comments and compare with the database. Based on themining, if the user comment is matched with the positive keywords in database then admin will recommend the productfor users. We use Hybrid filtering technique in recommendation system with feedback analysis to improve therecommendation system. These feedbacks include reviews, ratings and emoticons which are implemented by stochasticlearning algorithm to enhance the sentimental analysis of a product more precisely for customer. For feedback analysiswe build a dataset based on a survey of online mobile products comments from various shopping site. We alsoproposed to eliminate fake reviews of a product from the online site. This can be achieved by MAC address, where usercan post comments only one time for the purchased product along with MAC address. This helps the customer to comeout with genuine reviews of the product.
关键词：Recommendation System; Hybrid Filtering technique; Opinion Mining; Fake reviews; MAC address.