Digital platforms regulation - Next session

Speaker: Julien M’Barki - University Paris 1 Panthéon-Sorbonne

Paper : Algo-Rhythm Unplugged: Effects of Explaining Algorithmic Recommendations on Music Discovery

Abstract : Music streaming services make massive use of algorithms in their music recommender systems (MRS) to guide users to tracks that they are likely to enjoy. However, the black-box nature of these algorithms makes them difficult for users to understand, both in terms of how they work and the music they predict. The field of explainable AI (XAI), and in particular its "explanation" side, has emerged to make the uses of AI (including MRSs) more comprehensible to users. This paper aims to observe, using a laboratory experiment, whether the explanation of an MRS algorithm induces a change in discovery behavior on music streaming services. In a theoretical framework, we model two types of user discovery behavior, namely "study" and "browse" behaviors. We then test in the lab the explanation effects of these behaviors by explaining a simplified "semi-personalized" MRS and measuring the relative listening time of the tracks. Data shows no average effect induced by the explanations, but we observe a differentiated impact of explanations based on the treatment intensity (i.e. the time spent reading them). The more individuals are treated, the more they listen to the tracks, reinforcing the "study" behavior.

Relevant information:

• Here the Microsoft Teams link. The session will cover 90 minutes (questions and answers included).
• The working group is part of the activities of the Governance and Regulation Chair at Paris Dauphine-PSL University. The group meets once in two weeks to discuss, share, and present working papers and/or ideas surrounding recent developments in the domain of platform regulation. We have both internal and external presenters to share their work.
• If you would like to present your work or have anyone in mind to present, please send an email to lucas.eustache@dauphine.psl.eu.