Contextual Personalized Re-Ranking of Music Recommendations through Audio Features

TL;DR

This paper introduces a personalized re-ranking algorithm for music recommendations that leverages audio features and contextual information, demonstrating improved accuracy over global models using real-world Twitter data.

Contribution

The paper presents a novel personalized re-ranking method based on audio features that adapts to user-specific contexts, outperforming global models in music recommendation accuracy.

Findings

Personalized model outperforms global model in precision and MAP.

Audio features effectively capture user preferences in specific contexts.

Evaluation conducted on real-world Twitter-based dataset.

Abstract

Users are able to access millions of songs through music streaming services like Spotify, Pandora, and Deezer. Access to such large catalogs, created a need for relevant song recommendations. However, user preferences are highly subjective in nature and change according to context (e.g., music that is suitable in the morning is not as suitable in the evening). Moreover, the music one user may prefer in a given context may be different from what another user prefers in the same context (i.e., what is considered good morning music differs across users). Accurately representing these preferences is essential to creating accurate and effective song recommendations. User preferences for songs can be based on high level audio features, such as tempo and valence. In this paper, we therefore propose a contextual re-ranking algorithm, based on audio feature representations of user preferences in specific contextual conditions. We evaluate the performance of our re-ranking algorithm using the #NowPlaying-RS dataset, which exists of user listening events crawled from Twitter and is enriched with song audio features. We compare a global (context for all users) and personalized (context for each user) model based on these audio features. The global model creates an audio feature representation of each contextual condition based on the preferences of all users. Unlike the global model, the personalized model creates user-specific audio feature representations of contextual conditions, and is measured across 333 distinct users. We show that the personalized model outperforms the global model when evaluated using the precision and mean average precision metrics.