Biases in Automated Music Playlist Generation: A Comparison of Next-Track Recommending Techniques

Dietmar Jannach 1 Iman Kamehkhosh 1 Geoffray Bonnin 2
2 KIWI - Knowledge Information and Web Intelligence
LORIA - Laboratoire Lorrain de Recherche en Informatique et ses Applications
Abstract : Playlist generation is a special form of music recommendation where the problem is to create a sequence of tracks to be played next, given a number of seed tracks. In academia, the evaluation of playlisting techniques is often done by assessing with the help of Information Retrieval measures if an algorithm is capable of selecting those tracks that also a human would pick next. Such approaches however cannot capture other factors, e.g., the homogeneity of the tracks that can determine the quality perception of playlists. In this work, we report the results of a multi-metric comparison of different academic approaches and a commercial playlisting service. Our results show that all tested techniques generate playlists with certain biases, e.g., towards very popular tracks, and often create playlists continuations that are quite different from those that are created by real users.
Type de document :
Communication dans un congrès
User Modeling, Adaptation and Personalization, Jul 2016, Halifax, Canada
Liste complète des métadonnées

https://hal.inria.fr/hal-01305105
Contributeur : Geoffray Bonnin <>
Soumis le : mercredi 20 avril 2016 - 16:42:29
Dernière modification le : mardi 24 avril 2018 - 13:36:55

Identifiants

  • HAL Id : hal-01305105, version 1

Collections

Citation

Dietmar Jannach, Iman Kamehkhosh, Geoffray Bonnin. Biases in Automated Music Playlist Generation: A Comparison of Next-Track Recommending Techniques. User Modeling, Adaptation and Personalization, Jul 2016, Halifax, Canada. 〈hal-01305105〉

Partager

Métriques

Consultations de la notice

141