Survey on Recommender System using Distributed Framework

Recommender systems form a specific type of In-formation Filtering (IF) technique. It makes use of content based, collaborative filtering or hybrid approach. In this paper, all the approaches of recommender system is explained. Classification of data is vital part of recommender system. The Bayesian Classifier is one of the most successful machine learning algorithms in many classification domains. As scale of recommender system continues to expand, the number of items and users of recommender system is growing exponentially. The impact of this, the single node machine implementing these algorithms is time consuming and unable to meet the computing needs of large data sets. To improve the performance distributed processing of big data across multiple clusters of nodes is needed. Apache hadoop is an parallel distributed framework. Hadoop distributed file system allows distributed processing of big data across multiple clusters of nodes. The recommender input data will be encapsulated in Apache Mahout and enhance efficiency of recommendation using parallel distributed framework.