RECOMMENDATION SYSTEM USING KEYWORD BASE APPROACH

Abstract: Big data refers to datasets that aren't solely massive, however additionally high in variety and velocity that makes them troublesome to handle using tradition tools and techniques. Due to the rise of such Data, solutions got to be studied and provided so as to handle and extract worth and information from these datasets. Nowadays Web services are very widespread .Recommender systems represent user preferences for the aim of suggesting things to get or examine. They are many basic applications in electronic commerce and data access, providing suggestions that effectively prune massive data areas so users are directed toward those things that best meet their wants and preferences. A variety of techniques are projected for activity recommendation, including content-based, collaborative, knowledge-based and different techniques. In this paper, we are presenting "Keyword-Aware Service Recommendation Method", to deal with the above challenges. It aims at presenting a customized service recommendation list and recommending the foremost applicable services to the users effectively. Specifically, keywords area unit wont to indicate users' preferences, and a user-based cooperative Filtering algorithm is adopted to get applicable recommendations. To improve the scalability and efficiency of KASR in "Big Data" environment, the proposed system proposes techniques that have been implemented it on a Map Reduce framework in Hadoop platform.  

Keywords: Recommender system, Preference, Keyword, Big data, Hadoop, MapReduce.