Improving automatic music tag annotation using stacked generalization of probabilistic SVM outputs

Steven R. Ness, Anthony Theocharis, George Tzanetakis, Luis Gustavo Martins

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

77 Citations (Scopus)

Abstract

Music listeners frequently use words to describe music. Personalized music recommendation systems such as Last.fm and Pandora rely on manual annotations (tags) as a mechanism for querying and navigating large music collections. A well-known issue in such recommendation systems is known as the cold-start problem: it is not possible to recommend new songs/tracks until those songs/tracks have been manually annotated. Automatic tag annotation based on content analysis is a potential solution to this problem and has recently been gaining attention. We describe how stacked generalization can be used to improve the performance of a state-of-the-art automatic tag annotation system for music based on audio content analysis and report results on two publicly available datasets.
Original languageEnglish
Title of host publicationMM'09 - Proceedings of the 2009 ACM Multimedia Conference, with Co-located Workshops and Symposiums
Pages705-708
Number of pages4
DOIs
Publication statusPublished - 2009
Event17th ACM International Conference on Multimedia, MM'09, with Co-located Workshops and Symposiums - Beijing, China
Duration: 19 Oct 200924 Oct 2009

Publication series

NameMM'09 - Proceedings of the 2009 ACM Multimedia Conference, with Co-located Workshops and Symposiums

Conference

Conference17th ACM International Conference on Multimedia, MM'09, with Co-located Workshops and Symposiums
Country/TerritoryChina
CityBeijing
Period19/10/0924/10/09

Keywords

  • Folksonomies
  • Music information retrieval
  • Music recommendation
  • Sound analysis
  • Tags

Fingerprint

Dive into the research topics of 'Improving automatic music tag annotation using stacked generalization of probabilistic SVM outputs'. Together they form a unique fingerprint.

Cite this