Combining Sources of Description for Approximating Music Similarity Ratings

  • Daniel Wolff
  • Tillman Weyde
Conference paper

DOI: 10.1007/978-3-642-37425-8_9

Part of the Lecture Notes in Computer Science book series (LNCS, volume 7836)
Cite this paper as:
Wolff D., Weyde T. (2013) Combining Sources of Description for Approximating Music Similarity Ratings. In: Detyniecki M., García-Serrano A., Nürnberger A., Stober S. (eds) Adaptive Multimedia Retrieval. Large-Scale Multimedia Retrieval and Evaluation. AMR 2011. Lecture Notes in Computer Science, vol 7836. Springer, Berlin, Heidelberg

Abstract

In this paper, we compare the effectiveness of basic acoustic features and genre annotations when adapting a music similarity model to user ratings. We use the Metric Learning to Rank algorithm to learn a Mahalanobis metric from comparative similarity ratings in in the MagnaTagATune database. Using common formats for feature data, our approach can easily be transferred to other existing databases. Our results show that genre data allow more effective learning of a metric than simple audio features, but a combination of both feature sets clearly outperforms either individual set.

Keywords

Music Information Retrieval Music Recommendation Computational Modelling Music Similarity Music Perception 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Daniel Wolff
    • 1
  • Tillman Weyde
    • 1
  1. 1.Department of ComputingCity University LondonLondonUK

Personalised recommendations