Modurec: Recommender Systems with Feature and Time Modulation

TL;DR

Modurec is an autoencoder-based recommender system that integrates user and item features with time information to address cold start and concept shift, achieving state-of-the-art results on Movielens datasets.

Contribution

It introduces feature-wise modulation in autoencoders for recommender systems, effectively combining features and time to improve predictions under data scarcity and distribution changes.

Findings

Outperforms standard autoencoder-based methods on Movielens.

Effectively mitigates cold start and concept shift issues.

Achieves state-of-the-art results in most evaluated settings.

Abstract

Current state of the art algorithms for recommender systems are mainly based on collaborative filtering, which exploits user ratings to discover latent factors in the data. These algorithms unfortunately do not make effective use of other features, which can help solve two well identified problems of collaborative filtering: cold start (not enough data is available for new users or products) and concept shift (the distribution of ratings changes over time). To address these problems, we propose Modurec: an autoencoder-based method that combines all available information using the feature-wise modulation mechanism, which has demonstrated its effectiveness in several fields. While time information helps mitigate the effects of concept shift, the combination of user and item features improve prediction performance when little data is available. We show on Movielens datasets that these modifications produce state-of-the-art results in most evaluated settings compared with standard autoencoder-based methods and other collaborative filtering approaches.