Neural Representations in Hybrid Recommender Systems: Prediction versus Regularization

TL;DR

This paper introduces the neural representation for prediction (NRP) framework for autoencoder-based hybrid recommender systems, demonstrating that neural representations improve rating prediction accuracy over traditional methods.

Contribution

The paper proposes the NRP framework, providing theoretical analysis and applying it to a direct neural network model that outperforms state-of-the-art methods in prediction tasks.

Findings

Neural representations outperform traditional representations for rating prediction.

The NRP framework improves prediction accuracy and reduces training time.

The direct neural network structure with NRP outperforms existing methods.

Abstract

Autoencoder-based hybrid recommender systems have become popular recently because of their ability to learn user and item representations by reconstructing various information sources, including users' feedback on items (e.g., ratings) and side information of users and items (e.g., users' occupation and items' title). However, existing systems still use representations learned by matrix factorization (MF) to predict the rating, while using representations learned by neural networks as the regularizer. In this paper, we define the neural representation for prediction (NRP) framework and apply it to the autoencoder-based recommendation systems. We theoretically analyze how our objective function is related to the previous MF and autoencoder-based methods and explain what it means to use neural representations as the regularizer. We also apply the NRP framework to a direct neural network structure which predicts the ratings without reconstructing the user and item information. We conduct extensive experiments on two MovieLens datasets and two real-world e-commerce datasets. The results confirm that neural representations are better for prediction than regularization and show that the NRP framework, combined with the direct neural network structure, outperforms the state-of-the-art methods in the prediction task, with less training time and memory.