Movie Recommendations Using the Deep Learning Approach

Recommendation systems are an important part of suggesting items especially in streaming services. For streaming movie services like Netflix, recommendation systems are essential for helping users find new movies to enjoy. In this paper, we propose a deep learning approach based on autoencoders to produce a collaborative filtering system which predicts movie ratings for a user based on a large database of ratings from other users. Using the MovieLens dataset, we explore the use of deep learning to predict users' ratings on new movies, thereby enabling movie recommendations. To verify the novelty and accuracy of our deep learning approach, we compare our approach to standard collaborative filtering techniques: k-nearest-neighbor and matrix-factorization. The experimental results show that our recommendation system outperforms a user-based neighborhood baseline both in terms of root mean squared error on predicted ratings and in a survey in which users judge between recommendations from both systems.