Transformer-Empowered Content-Aware Collaborative Filtering

TL;DR

This paper introduces a novel Transformer-based content-aware collaborative filtering approach that integrates structured knowledge graph data and unstructured content features, demonstrating improved recommendation performance.

Contribution

It proposes a new model combining KG-enhanced meta-preference networks with Transformer-empowered content features and introduces a cross-system contrastive learning training scheme.

Findings

Outperforms baseline systems in recommendation accuracy

Enhances knowledge-graph-based collaborative filtering with Transformer features

Provides new large-scale datasets for recommendation research

Abstract

Knowledge graph (KG) based Collaborative Filtering is an effective approach to personalizing recommendation systems for relatively static domains such as movies and books, by leveraging structured information from KG to enrich both item and user representations. Motivated by the use of Transformers for understanding rich text in content-based filtering recommender systems, we propose Content-aware KG-enhanced Meta-preference Networks as a way to enhance collaborative filtering recommendation based on both structured information from KG as well as unstructured content features based on Transformer-empowered content-based filtering. To achieve this, we employ a novel training scheme, Cross-System Contrastive Learning, to address the inconsistency of the two very different systems and propose a powerful collaborative filtering model and a variant of the well-known NRMS system within this modeling framework. We also contribute to public domain resources through the creation of a large-scale movie-knowledge-graph dataset and an extension of the already public Amazon-Book dataset through incorporation of text descriptions crawled from external sources. We present experimental results showing that enhancing collaborative filtering with Transformer-based features derived from content-based filtering outperforms strong baseline systems, improving the ability of knowledge-graph-based collaborative filtering systems to exploit item content information.