A Hybrid Approach to Music Playlist Continuation Based on Playlist-Song Membership

TL;DR

This paper presents a hybrid music playlist continuation model based on playlist-song membership, capable of extending both profiled and non-profiled playlists, and recommending rare or unseen songs by analyzing feature vectors.

Contribution

The paper introduces a novel hybrid model that improves playlist continuation by leveraging playlist-song membership, overcoming limitations of pure collaborative filtering methods.

Findings

Performs comparably to state-of-the-art collaborative filtering on common tasks.

Can extend non-profiled playlists and recommend seldom or unseen songs.

Shows robustness in recommending rare or new songs.

Abstract

Automated music playlist continuation is a common task of music recommender systems, that generally consists in providing a fitting extension to a given playlist. Collaborative filtering models, that extract abstract patterns from curated music playlists, tend to provide better playlist continuations than content-based approaches. However, pure collaborative filtering models have at least one of the following limitations: (1) they can only extend playlists profiled at training time; (2) they misrepresent songs that occur in very few playlists. We introduce a novel hybrid playlist continuation model based on what we name "playlist-song membership", that is, whether a given playlist and a given song fit together. The proposed model regards any playlist-song pair exclusively in terms of feature vectors. In light of this information, and after having been trained on a collection of labeled playlist-song pairs, the proposed model decides whether a playlist-song pair fits together or not. Experimental results on two datasets of curated music playlists show that the proposed playlist continuation model compares to a state-of-the-art collaborative filtering model in the ideal situation of extending playlists profiled at training time and where songs occurred frequently in training playlists. In contrast to the collaborative filtering model, and as a result of its general understanding of the playlist-song pairs in terms of feature vectors, the proposed model is additionally able to (1) extend non-profiled playlists and (2) recommend songs that occurred seldom or never in training~playlists.