Recommender System with Machine Learning and Artificial Intelligence

13Netflix Challenge—Improving Movie Recommendations

Vasu Goel

Computer Science and Engineering, SRM University, Delhi-NCR Campus, India

Abstract

This analysis is based on the challenge that Netflix offered to the data science community. The challenge was to improve Netflix’s movie recommendation system by 10% [9]. The objective of this analysis is to train multiple machine learning models using inputs from one data set to predict movie ratings in another data set.

Keywords: Movie effect model, user effect model, residual mean square error (RMSE), naive approach, predicted rating, regularization

13.1 Introduction

The Netflix movie rating scheme uses a rating range from 0 to 5. The data set from which the two sets mentioned previously will be created can be found using the following link: http://files.grouplens.org/datasets/movielens/ml-10m.zip. Recommendation systems use ratings that users have given items to make specific recommendations [13]. Companies that sell many products to many customers and permit these customers to rate their products, like Amazon, are able to collect massive datasets that can be used to predict what rating a particular user will give a specific item. Items for which a high rating is predicted for a given user are then recommended to that user. Netflix uses a recommendation system to predict how many stars a user will give a specific movie. Based on the predictions, the movie with the highest predicted rating for a user is then recommended to the user. ...