This paper presents our recent initiatives to foster the discoverability of new releases on the music streaming service Deezer. After introducing our search and recommendation features dedicated to new releases, we outline our shift from editorial to personalized release suggestions using cold start embeddings and contextual bandits. Backed by online experiments, we discuss the advantages of this shift in terms of recommendation quality and exposure of new releases on the service.
@article{arxiv.2401.02827,
title = {Let's Get It Started: Fostering the Discoverability of New Releases on Deezer},
author = {Léa Briand and Théo Bontempelli and Walid Bendada and Mathieu Morlon and François Rigaud and Benjamin Chapus and Thomas Bouabça and Guillaume Salha-Galvan},
journal= {arXiv preprint arXiv:2401.02827},
year = {2024}
}
Comments
Accepted for presentation as an "Industry Talk" at the 46th European Conference on Information Retrieval (ECIR 2024)