<< Back

by Paul Lamere (Sun Labs) and Òscar Celma (Music Technology Group)



The original proposal, in PDF

As the world of online music grows, music recommendation systems become an increasingly important way for music listeners to discover new music. Commercial recommenders such as Last.fm and Pandora have enjoyed commercial and critical success. But how well do these systems really work? How good are the recommendations? How far into the "long tail" do these recommenders reach?

In this tutorial we look at the current state-of-the-art in music recommendation. We examine current commercial and research systems, focusing on the advantages and the disadvantages of the various recommendation strategies. We look at some of the challenges in building music recommenders and we explore some of the ways that Music Information Retrieval (MIR) techniques can be used to improve future recommenders.


This tutorial presents an overview of the music recommendation problem, describes the various issues and challenges in developing a music recommender, explores the various approaches used for recommendation, including a detailed discussion of the advantages and disadvantages of these approaches, gives real-world examples of different types of recommenders and finally presents some of the challenges to building the next generation of music recommenders.

This tutorial also presents results from a subjective evaluation of a number of music recommendation systems, as well as presenting results using objective measures -based on graph and complex networks theory- to compare inherent differences between the various recommendation techniques, including collaborative filtering and content-based methods.

The outline of the talk will be:

  1. Introduction: Why music recommendation is important?
  2. Formalization of the recommendation problem
  3. Problems with recommenders
  4. Types of recommenders
  5. Recommender examples
  6. Evaluation of recommender algorithms
  7. Conclusions and Future of music recommendation


Sunday, September 23rd. 2007, 15:00 - 18:00

TU Wien - Freihaus, Wiedner Hauptstrasse 8-10, 1040 Wien, Room HS6. ISMIR 2007


This tutorial will be of interest to the general ISMIR audience, especially to those who are interested in a deeper understanding of the current state-of-the-art of music recommender systems and especially to those who are looking to direct their research toward areas with high commercial interest.


Oscar Celma

Oscar Celma is a researcher at Music Technology Group since 2000, and Lecturer at the Pompeu Fabra University, Barcelona (Spain). The main focus of his research lies in the music recommendation arena, especially hybrid approaches.

Since 2006 he is an Invited Expert of the W3C Multimedia Semantics Incubator Group. He is a member of the program committee of the Workshop on Learning the Semantics of Audio Signals (LSAS).

In 2006, Oscar received the 2nd prize in the International Semantic Web Challenge for the system named "Foafing the Music", a personalized music recommendation that exploits music related information available from the web.

During his undergraduate studies, he also obtained the diplomas in classical guitar, and composition. Though, nowadays he only makes some noise from time to time, with his old Grestch G6118 Anniversary.

Paul Lamere

Paul Lamere is the Principal Investigator for a project called "Search Inside the Music" at Sun Labs where he explores new ways to help people find highly relevant music, even as music collections get very large. Paul is especially interested in hybrid music recommenders and using visualizations to aid music discovery.

Paul serves on the program committee for ISMIR 2007 as well as on the program committee for Recommenders'07. Paul also authors "Duke Listens!" a blog focusing on music discovery and recommendation.