Leveraging Artist Catalogs for Cold-Start Music Recommendation

TL;DR

This paper introduces ACARec, an attention-based model that leverages artist catalogs to improve cold-start music recommendation, significantly enhancing prediction accuracy for new tracks.

Contribution

It proposes a novel artist-aware approach that uses artist catalogs to generate embeddings for new tracks, outperforming content-only methods.

Findings

Artist-aware methods more than double Recall and NDCG compared to baselines.

ACARec effectively predicts user preferences for new tracks and artists.

The approach improves cold item popularity estimation.

Abstract

The item cold-start problem poses a fundamental challenge for music recommendation: newly added tracks lack the interaction history that collaborative filtering (CF) requires. Existing approaches often address this problem by learning mappings from content features such as audio, text, and metadata to the CF latent space. However, previous works either omit artist information or treat it as just another input modality, missing the fundamental hierarchy of artists and items. Since most new tracks come from artists with previous history available, we frame cold-start track recommendation as 'semi-cold' by leveraging the rich collaborative signal that exists at the artist level. We show that artist-aware methods can more than double Recall and NDCG compared to content-only baselines, and propose ACARec, an attention-based architecture that generates CF embeddings for new tracks by attending over the artist's existing catalog. We show that our approach has notable advantages in predicting user preferences for new tracks, especially for new artist discovery and more accurate estimation of cold item popularity.