# Automatic playlist continuation using a hybrid recommender system combining features from text and audio

## TL;DR

This paper presents a hybrid music recommendation system for playlist continuation that combines matrix factorization with audio and text features and track co-occurrence data, achieving competitive results in a challenge.

## Contribution

The paper introduces a novel hybrid recommender system integrating audio, text, and co-occurrence features for playlist continuation tasks.

## Key findings

- System ranked 4th in ACM RecSys Challenge 2018
- Combining models improves recommendation quality
- Efficient approach with good overall performance

## Abstract

The ACM RecSys Challenge 2018 focuses on music recommendation in the context of automatic playlist continuation. In this paper, we describe our approach to the problem and the final hybrid system that was submitted to the challenge by our team Cocoplaya. This system consists in combining the recommendations produced by two different models using ranking fusion. The first model is based on Matrix Factorization and it incorporates information from tracks' audio and playlist titles. The second model generates recommendations based on typical track co-occurrences considering their proximity in the playlists. The proposed approach is efficient and achieves a good overall performance, with our model ranked 4th on the creative track of the challenge leaderboard.