An Experimental Comparison of Similarity Adaptation Approaches

  • Sebastian Stober
  • Andreas Nürnberger
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7836)


Similarity plays an important role in many multimedia retrieval applications. However, it often has many facets and its perception is highly subjective – very much depending on a person’s background or retrieval goal. In previous work, we have developed various approaches for modeling and learning individual distance measures as a weighted linear combination of multiple facets in different application scenarios. Based on a generalized view of these approaches as an optimization problem guided by generic relative distance constraints, we describe ways to address the problem of constraint violations and finally compare the different approaches against each other. To this end, a comprehensive experiment using the Magnatagatune benchmark dataset is conducted.


Constraint Violation Distance Constraint Similarity Judgment Weighted Linear Combination Similarity Adaptation 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Bade, K.: Personalized Hierarchical Structuring. PhD thesis, Otto-von-Guericke-University Magdeburg (2009)Google Scholar
  2. 2.
    Bade, K., Garbers, J., Stober, S., Wiering, F., Nürnberger, A.: Supporting folk-song research by automatic metric learning and ranking. In: Proc. of the 10th Int. Conf. on Music Information Retrieval (ISMIR 2009) (2009)Google Scholar
  3. 3.
    Bade, K., Nürnberger, A.: Creating a cluster hierarchy under constraints of a partially known hierarchy. In: Proc. of the 2008 SIAM Int. Conf. on Data Mining (2008)Google Scholar
  4. 4.
    Cheng, W., Hüllermeier, E.: Learning similarity functions from qualitative feedback. In: Althoff, K.-D., Bergmann, R., Minor, M., Hanft, A. (eds.) ECCBR 2008. LNCS (LNAI), vol. 5239, pp. 120–134. Springer, Heidelberg (2008)CrossRefGoogle Scholar
  5. 5.
    Donaldson, J., Lamere, P.: Using visualizations for music discovery. In: Tutorial at the 10th Int. Conf. on Music Information Retrieval (ISMIR 2009) (2009)Google Scholar
  6. 6.
    Fan, R., Chang, K., Hsieh, C., Wang, X., Lin, C.: Liblinear: A library for large linear classification. The Journal of Machine Learning Research 9, 1871–1874 (2008)zbMATHGoogle Scholar
  7. 7.
    Goldfarb, D., Idnani, A.: A numerically stable dual method for solving strictly convex quadratic programs. Mathematical Programming 27(1), 1–33 (1983)MathSciNetzbMATHCrossRefGoogle Scholar
  8. 8.
    Joachims, T.: A Support Vector Method for Multivariate Performance Measures. In: Proc. of the Int. Conf. on Machine Learning (ICML 2005) (2005)Google Scholar
  9. 9.
    Law, E., von Ahn, L.: Input-agreement: a new mechanism for collecting data using human computation games. In: Proc. of the 27th Int. Conf. on Human Factors in Computing Systems (CHI 2009) (2009)Google Scholar
  10. 10.
    Lübbers, D., Jarke, M.: Adaptive multimodal exploration of music collections. In: Proc. of the 10th Int. Conf. on Music Information Retrieval (ISMIR 2009) (2009)Google Scholar
  11. 11.
    McFee, B., Barrington, L., Lanckriet, G.: Learning similarity from collaborative filters. In: Proc. of the 11th Int. Conf. on Music Information Retrieval (ISMIR 2010) (2010)Google Scholar
  12. 12.
    McFee, B., Lanckriet, G.: Heterogeneous embedding for subjective artist similarity. In: Proc. of the 10th Int. Conf. on Music Information Retrieval (ISMIR 2009) (2009)Google Scholar
  13. 13.
    Nürnberger, A., Klose, A.: Improving clustering and visualization of multimedia data using interactive user feedback. In: Proc. of the 9th Int. Conf. on Information Processing and Management of Uncertainty in Knowledge-Based Systems (IPMU 2002) (2002)Google Scholar
  14. 14.
    Salton, G., Buckley, C.: Term weighting approaches in automatic text retrieval. Information Processing & Management 24(5), 513–523 (1988)CrossRefGoogle Scholar
  15. 15.
    Slaney, M., Weinberger, K.Q., White, W.: Learning a metric for music similarity. In: Proc. of the 9th Int. Conf. on Music Information Retrieval (ISMIR 2008) (2008)Google Scholar
  16. 16.
    Nürnberger, A., Stober, S.: User modelling for interactive user-adaptive collection structuring. In: Boujemaa, N., Detyniecki, M., Nürnberger, A. (eds.) AMR 2007. LNCS, vol. 4918, pp. 95–108. Springer, Heidelberg (2008)CrossRefGoogle Scholar
  17. 17.
    Stober, S., Nürnberger, A.: Towards user-adaptive structuring and organization of music collections. In: Detyniecki, M., Leiner, U., Nürnberger, A. (eds.) AMR 2008. LNCS, vol. 5811, pp. 53–65. Springer, Heidelberg (2010)CrossRefGoogle Scholar
  18. 18.
    Stober, S., Nürnberger, A.: Similarity adaptation in an exploratory retrieval scenario. In: Detyniecki, M., Knees, P., Nürnberger, A., Schedl, M., Stober, S. (eds.) AMR 2010. LNCS, vol. 6817, pp. 144–158. Springer, Heidelberg (2012)CrossRefGoogle Scholar
  19. 19.
    Stober, S.: Adaptive distance measures for exploration and structuring of music collections. In: Proc. of AES 42nd Conf. on Semantic Audio (2011)Google Scholar
  20. 20.
    Wolff, D., Weyde, T.: Combining Sources of Description for Approximating Music Similarity Ratings. In: Detyniecki, M., García-Serrano, A., Nürnberger, A., Stober, S. (eds.) AMR 2011. LNCS, vol. 7836, pp. 114–124. Springer, Heidelberg (2013)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2013

Authors and Affiliations

  • Sebastian Stober
    • 1
  • Andreas Nürnberger
    • 1
  1. 1.Data & Knowledge Engineering Group, Faculty of Computer ScienceOtto-von-Guericke-University MagdeburgMagdeburgGermany

Personalised recommendations