Abstract: Service recommendation system is an one of the valuable tool and it provide a appropriate recommendation to the users. Now a days large number of customers, services and the other online information needs service recommendation system. It provides the recommendations about the services to the user. In the existing recommendation system, the ratings of services and the service recommendation lists presented to users are the same.In the existing system the major problem is scalability and inefficiency problems when processing or analyzing such large-scale data. Here we propose a Keyword-Aware Service Recommendation method. In the proposed system it recommends the services based on the user preferences. In this system keyword refers to the user preferences. In this system user based collaborative filtering algorithm is used to generate recommendations. To improve the efficiency here we implement the system in Hadoop. In addition to that to make services effective, here we are going to use rank boosting algorithm with combined preferences. To achieve the scalability and efficiency with help of mapreduce framework in an big data environment.

Keywords: Recommendation system, Hadoop, collaborative algorithm, preference, keyword, Mapreduce.