Recommending Movies Using User-Based and Item-Based Collaborative Filtering on R Platform

Abstract

Recommender systems are information filtering systems that deal with the problem of information overload by filtering vital information fragment out of a large amount of dynamically generated information according to user’s preferences, interest or observed behaviour about an item. Recommender system has the ability to predict whether a particular user would prefer an item or not based on the user’s profile. Recommendation algorithms are best known for their use on e-commerce websites, where they use inputs about a customer’s interests to generate a list of recommended items. A new emerging sector in India, online movie viewership and subscription of movies, demands an expert and highly technical skill to understand the movie preferences of the viewers spread across India. This paper uses two techniques (run on R platform), user-based collaborative filtering and item-based collaborative filtering, to understand the preferences of people (without giving any reason to it) and recommending mechanism was solely based upon user-user similarity matrix and item-item similarity matrix. A dataset of 563 movies and 9,985 consumers from Amazon prime has been taken for recommending movies for people who have not watched a particular set of movies. The robustness of the two techniques is also compared and explained.

Keywords: User-Based Collaborative Filtering, Item-Based Collaborative Filtering, Recommendation Engine

View PDF