Skip prediction using boosting trees based on acoustic features of tracks in sessions

TL;DR

This paper presents a boosting trees-based system for predicting track skips in sessions, using acoustic features, achieving competitive performance in the Spotify challenge.

Contribution

It introduces a novel ensemble approach combining multiple boosting trees trained on session and track features for skip prediction.

Findings

Achieved a MAA of 0.554 on the challenge dataset

Ranked 14th out of over 600 submissions

Demonstrated effectiveness of acoustic feature-based models

Abstract

The Spotify Sequential Skip Prediction Challenge focuses on predicting if a track in a session will be skipped by the user or not. In this paper, we describe our approach to this problem and the final system that was submitted to the challenge by our team from the Music Technology Group (MTG) under the name "aferraro". This system consists in combining the predictions of multiple boosting trees models trained with features extracted from the sessions and the tracks. The proposed approach achieves good overall performance (MAA of 0.554), with our model ranked 14th out of more than 600 submissions in the final leaderboard.