Collaborative Filtering with Recurrent Neural Networks

TL;DR

This paper presents a novel approach to collaborative filtering by framing it as a sequence prediction task and applying LSTM neural networks, demonstrating competitive performance and advantages over traditional methods.

Contribution

It introduces the application of recurrent neural networks, specifically LSTM, to collaborative filtering, highlighting their effectiveness in recommendation tasks.

Findings

LSTM-based collaborative filtering is competitive with traditional methods.

LSTM outperforms in item coverage and short-term predictions.

Recurrent neural networks can effectively model user-item interactions.

Abstract

We show that collaborative filtering can be viewed as a sequence prediction problem, and that given this interpretation, recurrent neural networks offer very competitive approach. In particular we study how the long short-term memory (LSTM) can be applied to collaborative filtering, and how it compares to standard nearest neighbors and matrix factorization methods on movie recommendation. We show that the LSTM is competitive in all aspects, and largely outperforms other methods in terms of item coverage and short term predictions.