A Deep Multimodal Approach for Cold-start Music Recommendation
This addresses the challenge of recommending new artists with scarce prior knowledge for music streaming services, though it is incremental in nature.
The paper tackles the cold-start problem in music recommendation by combining text, audio, and user feedback data using a deep multimodal network, resulting in improved recommendation accuracy.
An increasing amount of digital music is being published daily. Music streaming services often ingest all available music, but this poses a challenge: how to recommend new artists for which prior knowledge is scarce? In this work we aim to address this so-called cold-start problem by combining text and audio information with user feedback data using deep network architectures. Our method is divided into three steps. First, artist embeddings are learned from biographies by combining semantics, text features, and aggregated usage data. Second, track embeddings are learned from the audio signal and available feedback data. Finally, artist and track embeddings are combined in a multimodal network. Results suggest that both splitting the recommendation problem between feature levels (i.e., artist metadata and audio track), and merging feature embeddings in a multimodal approach improve the accuracy of the recommendations.