Order, context and popularity bias in next-song recommendations

Andreu Vall,Massimo Quadrana,Gerhard Widmer,Markus Schedl

doi:10.1007/s13735-019-00169-8

Abstract

The availability of increasingly larger multimedia collections has fostered extensive research in recommender systems. Instead of capturing general user preferences, the task of next-item recommendation focuses on revealing specific session preferences encoded in the most recent user interactions. This study focuses on the music domain, particularly on the task of music playlist continuation, a paradigmatic case of next-item recommendation. While the accuracy achieved in next-song recommendations is important, in this work we shift our focus toward a deeper understanding of fundamental playlist characteristics, namely the song order, the song context and the song popularity, and their relation to the recommendation of playlist continuations. We also propose an approach to assess the quality of the recommendations that mitigates known problems of off-line experiments for music recommender systems. Our results indicate that knowing a longer song context has a positive impact on next-song recommendations. We find that the long-tailed nature of the playlist datasets makes simple and highly expressive playlist models appear to perform comparably, but further analysis reveals the advantage of using highly expressive models. Finally, our experiments suggest that the song order is not crucial to accurately predict next-song recommendations.

Highlights

Automated music playlist continuation is a specific task in music recommender systems where the user sequentially receives song recommendations, producing a listening experience similar to traditional radio broadcasting
We have explicitly investigated the impact of the song order, the song context and the popularity bias in music playlists for the task of predicting next-song recommendations
Our results indicate that the playlist-based collaborative filtering (CF) model and the recurrent neural networks (RNNs) model, which can consider the full song context, do benefit from increasingly longer song contexts

Summary

Introduction

Automated music playlist continuation is a specific task in music recommender systems where the user sequentially receives song recommendations, producing a listening experience similar to traditional radio broadcasting. Sequential recommendation scenarios are very natural in the music domain. This is possibly explained by the short time required to listen to a song, which results in listening sessions typically including not one, but several songs. We refer to the current and previous songs in a playlist as the “song context” available to the recommender system when it predicts the song. This terminology is borrowed from language models and should not be confused with the incorporation of user’s contextual information into the recommender system

Objectives

Results

Conclusion