On Euclidean Corrections for Non-Euclidean Dissimilarities

  • Robert P. W. Duin
  • Elżbieta Pękalska
  • Artsiom Harol
  • Wan-Jui Lee
  • Horst Bunke
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5342)


Non-Euclidean dissimilarity measures can be well suited for building representation spaces that are more beneficial for pattern classification systems than the related Euclidean ones [1,2]. A non-Euclidean representation space is however cumbersome for training classifiers, as many statistical techniques rely on the Euclidean inner product that is missing there. In this paper we report our findings on the applicability of corrections that transform a non-Euclidean representation space into a Euclidean one in which similar or better classifiers can be trained. In a case-study based on four principally different classifiers we find out that standard correction procedures fail to construct an appropriate Euclidean space, equivalent to the original non-Euclidean one.


Negative Eigenvalue Edit Distance Dissimilarity Matrix Quadratic Discriminant Analysis Pattern Recognition Letter 
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.


  1. 1.
    Pekalska, E., Duin, R., Günter, S., Bunke, H.: On not making dissimilarities Euclidean. In: S+SSPR, pp. 1145–1154 (2004)Google Scholar
  2. 2.
    Pekalska, E., Harol, A., Duin, R., Spillmann, B., Bunke, H.: Non-Euclidean or non-metric measures can be informative. In: S+SSPR, pp. 871–880 (2006)Google Scholar
  3. 3.
    Bunke, H., Sanfeliu, A. (eds.): Syntactic and Structural Pattern Recognition Theory and Applications. World Scientific, Singapore (1990)zbMATHGoogle Scholar
  4. 4.
    Bunke, H., Shearer, K.: A graph distance metric based on the maximal common subgraph. Pattern Recognition Letters 19, 255–259 (1998)CrossRefzbMATHGoogle Scholar
  5. 5.
    Torsello, A., Hancock, E.: Computing approximate tree edit distance using relaxation labeling. Pattern Recognition Letters 24, 1089–1097 (2003)CrossRefzbMATHGoogle Scholar
  6. 6.
    Robles-Kelly, A., Hancock, E.: String edit distance, random walks and graph-matching. IJPRAI 18, 315–327 (2004)zbMATHGoogle Scholar
  7. 7.
    Pekalska, E., Duin, R.: The Dissimilarity Representation for Pattern Recognition. Foundations and Applications. World Scientific, Singapore (2005)Google Scholar
  8. 8.
    Wilson, R., Luo, B., Hancock, E.: Pattern vectors from algebraic graph theory. IEEE Trans. on PAMI 27, 1112–1124 (2005)CrossRefGoogle Scholar
  9. 9.
    Jacobs, D., Weinshall, D., Gdalyahu, Y.: Classification with Non-Metric Distances: Image Retrieval and Class Representation. IEEE TPAMI 22, 583–600 (2000)CrossRefGoogle Scholar
  10. 10.
    Haasdonk, B.: Feature space interpretation of SVMs with indefinite kernels. IEEE TPAMI 25, 482–492 (2005)CrossRefGoogle Scholar
  11. 11.
    Muñoz, A., de Diego, I.M.: From indefinite to positive semi-definite matrices. In: SSPR/SPR, pp. 764–772 (2006)Google Scholar
  12. 12.
    Andreu, G., Crespo, A., Valiente, J.M.: Selecting the toroidal self-organizing feature maps (TSOFM) best organized to object recogn. In: ICNN, pp. 1341–1346 (1997)Google Scholar
  13. 13.
    Goldfarb, L.: A new approach to pattern recognition. In: Kanal, L., Rosenfeld, A. (eds.) Progress in Pattern Recognition, vol. 2, pp. 241–402. Elsevier, Amsterdam (1985)Google Scholar
  14. 14.
    Gower, J.: Metric and Euclidean Properties of Dissimilarity Coefficients. J. of Classification 3, 5–48 (1986)MathSciNetCrossRefzbMATHGoogle Scholar
  15. 15.
    Chung, F.: Spectral Graph Theory. In: CBMS Regional Conference Series in Mathematics, vol. 92. American Mathematical Society (1997)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2008

Authors and Affiliations

  • Robert P. W. Duin
    • 1
  • Elżbieta Pękalska
    • 2
  • Artsiom Harol
    • 1
  • Wan-Jui Lee
    • 1
  • Horst Bunke
    • 3
  1. 1.Faculty of Electrical Engineering, Mathematics and Computer SciencesDelft University of TechnologyThe Netherlands
  2. 2.School of Computer ScienceUniversity of ManchesterUnited Kingdom
  3. 3.Department of Computer ScienceUniversity of BernSwitzerland

Personalised recommendations