Abstract: There is an extensive class of web applications that involve predicting user responses to options. Such a facility is called a recommendation system. In today’s world recommender’s system plays very vital role in recommending user required stuffs from the systems like shopping carts, hotel websites, tours and travels website, social media’s and other entertainment media websites. We aim to build such a distributed recommendation system on top of the Hadoop framework. Since the data to be processed is usually massive and the computational workload will become heavier after adding and then returning rating parameters, optimizations specific to this modified algorithm and also the Hadoop framework will be the major concern of our project. Collaborative filtering algorithm is one of the most popularly used algorithm for recommendations systems, since Collaborative filtering computational complexity is very high thus it is too much time consuming to use this for large scale data. In this paper we implement Collaborative filtering for the map-reduced recipe data in cloud computing, we use Hadoop to process the large recipe data and then use Map-Reduced data to publish the recipe to users and based on the user interest and rating we recommending the recipe items to the users.
Keywords: Collaborative Filtering, Recommender System, Hadoop, Cloud Computing, Map-Reduce.