Provision of Content Based Service Recommendations using Hadoop and MapReduce

Service recommender systems are widely known as the valuable tools for providing appropriate recommendations to users. Most of the existing service recommender systems give the same ratings and rankings of services to different users without considering diverse users' preferences, therefore it fail to meet users' personalized requirements. In recent years the amount of customers, services and online information has grown rapidly, yielding the big data analysis problem for service recommender systems. Traditional service recommender systems face performance issues like scalability and inefficiency problems when processing or analyzing such large-scale data. Here, we propose a method to address the above issues which aims at pre sending a personalized service recommendation list and recommending the most appropriate services to the users with different perspective. Specifically, s are the most helpful things used to indicate users' diverse preferences, and a user-based Collaborative Filtering algorithm is used to generate appropriate recommendations. To improve scalability and efficiency in a big data environment, the proposed system is implemented on Hadoop platform, which is a widely-adopted distributed computing platform using the MapReduce parallel processing paradigm. Hence, the proposed system significantly improves the accuracy and scalability of service recommender systems and provides the recommendations to the users efficiently.