A Hybrid Latent Variable Neural Network Model for Item Recommendation

TL;DR

This paper introduces a hybrid neural network model with latent variables that improves item recommendation accuracy and addresses the cold-start problem by integrating collaborative filtering with additional item or user information.

Contribution

The paper proposes a novel latent neural network model that combines collaborative filtering with content information to effectively solve the cold-start problem in recommendations.

Findings

LNN outperforms traditional content-based filters.

LNN matches the accuracy of state-of-the-art collaborative filtering.

LNN effectively mitigates the cold-start problem.

Abstract

Collaborative filtering is used to recommend items to a user without requiring a knowledge of the item itself and tends to outperform other techniques. However, collaborative filtering suffers from the cold-start problem, which occurs when an item has not yet been rated or a user has not rated any items. Incorporating additional information, such as item or user descriptions, into collaborative filtering can address the cold-start problem. In this paper, we present a neural network model with latent input variables (latent neural network or LNN) as a hybrid collaborative filtering technique that addresses the cold-start problem. LNN outperforms a broad selection of content-based filters (which make recommendations based on item descriptions) and other hybrid approaches while maintaining the accuracy of state-of-the-art collaborative filtering techniques.