A RECOMMENDER SYSTEM FOR ONLINE SOCIAL NETWORK USING HADLOOP IN LARGE SCALE USER GENERATED DATA

Abstract

We are living in an age of Data and Information. Online social networks are contributing in enlargement of this data on high scale and Recommendation systems are helping industries to make this data useful for business purposes. It is helping to enhance the opportunities in online social data. Online social network generate large quantity of data from its users and recommendation system use this data for suggesting right piece of information to the user. But in the time of Big Data, processing large volumes of data generating suggestions is a difficult job. We are aiming to implement a combined approach for recommendation algorithm which include user-based collaborative filtering and item-based collaborative filtering using Apache Mahout, a machine learning tool, on Hadoop platform to reduced the time for recommendation generation and to provide a scalable system for processing large data sets efficiently.