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：2015
卷号：3
期号：7
DOI：10.15680/ijircce.2015.03071138
出版社：S&S Publications
摘要：Various organizations need Decision Making System for enhancing the service providing ability towardsuser. So, the user behaviour analysis is required for decision making system to deal with understanding requirement ofusers. Also to provide efficient services through number of domain like product based system, reservation systems andmobile system etc. The analysis includes extraction of data related to various domains like ratings, rankings, reviews,post or any other information which is related to users feedback and product or system feature. In this paper HotelReservation System is used to provide personalized list of hotels to user from review processing by mining data interms of keywords extraction. The keywords are nothing but user preferences. Then prediction algorithm calledKeyword Aware Service Recommendation method is used to predict the behaviour of user based on occurrence ofkeywords in reviews. This is user-based collaborative filtering algorithm used for prediction which determines thesimilarity of set of keywords referred by current user and previous user. As the review analysis has big amount of data,it is processed using Hadoop environment and Map Reduce parallel programming framework to enhance the scalabilityand efficiency of KASR method. Finally experimental results are achieved based on hotel reviews dataset used fordemonstration which will improve KASR for accurate and scalable user behaviour system.
关键词：user preferences; keywords; recommendation; big data; Map Reduce
