Department of Computational Perception
Department of
Computational Perception
Johannes Kepler Universit+AOQ-t Linz


Home  –  Mission  –  Teaching  –  People  –  Research  –  Media  –  Impressum


PERSONALIZED MUSIC RETRIEVAL VIA MUSIC CONTENT, MUSIC CONTEXT, AND USER CONTEXT

Project Title: Personalized Music Retrieval via Music Content, Music Context, and User Context

Sponsor: Austrian Science Fund (Fonds zur Förderung der wissenschaftlichen Forschung, FWF)

Project Number: P22856-N23

Duration: 48 months (Nov 2010 Oct 2014)

Persons involved:

Markus Schedl (Project Leader)
Sebastian Böck
David Hauger

Abstract

One of the biggest challenges for today's users of digital music collections is how to find music that matches his or her taste depending on factors like the music content, the user's general music taste, but also his or her current emotional state, activity, situation, or surrounding. This project aims at supporting the user by modeling such aspects in terms of similarity functions, which will eventually be integrated to build a system for personalized access to music collections.

Aspects of music similarity can be broadly categorized into music content (e.g., the rhythm or timbre of a song), music context (e.g., song lyrics or terms used on an artist's Web page to describe his or her music), and user context (i.e., external human factors that may influence how a listener perceives music, for example, the listener's mood, location, used playback device, or listening situation).

In this project, we will address the third pillar of music similarity, i.e., taking into account the user context when he or she is listening to music. We are thus going to develop novel models to describe user-related aspects of music similarity, novel methods to combine the three broad dimensions of music similarity, and novel strategies to use this combined model for information retrieval (IR) tasks. These IR tasks will include, for example, a personalized music recommendation engine, user-adaptive playlist generation or an adaptive user interface to access music collections.

Facing the challenges described above, we will carry out research to accomplish the following six goals:

The expected benefits from the projects will be:

Publications