A Long-Tail Friendly Representation Framework for Artist and Music Similarity

TL;DR

This paper introduces a neural network-based framework designed to improve artist and music similarity representations, especially in long-tail scenarios with sparse data, enhancing music recommendation accuracy.

Contribution

It proposes the Long-Tail Friendly Representation Framework (LTFRF) that integrates multiple data types and uses a novel Multi-Relationship Loss to outperform GNNs in long-tail music recommendation tasks.

Findings

Outperforms baseline by 9.69% in Hit Ratio@10 for similar artist recommendation.

Achieves 19.42% higher in Hit Ratio@10 for music recommendation.

Improves long-tail recommendation metrics by over 11%.

Abstract

The investigation of the similarity between artists and music is crucial in music retrieval and recommendation, and addressing the challenge of the long-tail phenomenon is increasingly important. This paper proposes a Long-Tail Friendly Representation Framework (LTFRF) that utilizes neural networks to model the similarity relationship. Our approach integrates music, user, metadata, and relationship data into a unified metric learning framework, and employs a meta-consistency relationship as a regular term to introduce the Multi-Relationship Loss. Compared to the Graph Neural Network (GNN), our proposed framework improves the representation performance in long-tail scenarios, which are characterized by sparse relationships between artists and music. We conduct experiments and analysis on the AllMusic dataset, and the results demonstrate that our framework provides a favorable generalization of artist and music representation. Specifically, on similar artist/music recommendation tasks, the LTFRF outperforms the baseline by 9.69%/19.42% in Hit Ratio@10, and in long-tail cases, the framework achieves 11.05%/14.14% higher than the baseline in Consistent@10.