Enhancing Movie Recommender System

Abstract

Recommender system helps customers buying products/items efficiently and at the same time benefits the business. It can be built using approaches like: (1) Collaborative Filtering (2) Content Based Filtering and (3) Hybrid Filtering. In Collaborative Recommender System, ratings of the most similar users (in case of user based collaborative filtering) or items (in case of item based collaborative filtering) are used to predict the rating of the new item. In Content Based Filtering, user profile is constructed based on the contentof theitems liked by the user in the past and then based on similarity between user and item profile, recommendations are made. Hybrid Filtering combines collaborative and content based approach. In this paper, we focus on movie recommendation task. Prediction task is modelled as classification task where our aim is to predict whether the item (movie in our case) will be liked or disliked by the user. In our work, we propose an item based recommender which combines usage, tag and movie specific data such as genres, star castand directors to improve the accuracy of the Recommender System. We have tested ourapproach using Hetrec2011-movielens-2kdataset. We use Accuracy and F-measure to evaluate the performance of our proposed system.