Let's Get It Started (Again!): Fostering the Discoverability of New Releases on Deezer Léa Briand 0 Théo Bontempelli 0 Walid Bendada 0 1 Mathieu Morlon 0 François Rigaud 0 Benjamin Chapus 0 Thomas Bouabça 0 Guillaume Salha-Galvan 0 Deezer Research , Paris , France LAMSADE, Université Paris Dauphine, PSL , Paris , France

This workshop paper presents an extended version of a work previously been accepted for oral presentation as an Industry Talk at the 46th European Conference on Information Retrieval (ECIR 2024), under the title “Let's Get It Started: Fostering the Discoverability of New Releases on Deezer”. This conference version received the ECIR 2024 Best Industry Track Paper award. We present 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.

 eol>New Releases Music Recommendation Music Discovery Music Streaming Service A/B Testing