Abstract

In the process of designing pattern recognition systems one may choose a representation based on pairwise dissimilarities between objects. This is especially appealing when a set of discriminative features is difficult to find. Various classification systems have been studied for such a dissimilarity representation: the direct use of the nearest neighbor rule, the postulation of a dissimilarity space and an embedding to a virtual, underlying feature vector space.

It appears in several applications that the dissimilarity measures constructed by experts tend to have a non-Euclidean behavior. In this paper we first analyze the causes of such choices and then experimentally verify that the non-Euclidean property of the measure can be informative.

Keywords

Template Match Dissimilarity Measure Dissimilarity Matrix Neighbor Rule Dissimilarity Matrice 
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.

References

  1. 1.
    Edelman, S.: Representation and Recognition in Vision. MIT Press, Cambridge (1999)Google Scholar
  2. 2.
    Pękalska, E., Duin, R.: The Dissimilarity Representation for Pattern Recognition. In: Foundations and Applications. World Scientific, Singapore (2005)Google Scholar
  3. 3.
    Pękalska, E., Duin, R., Paclík, P.: Prototype selection for dissimilarity-based classifiers. Pattern Recognition 39(2), 189–208 (2006)MATHCrossRefGoogle Scholar
  4. 4.
    Dubuisson, M., Jain, A.: Modified Hausdorff distance for object matching. In: Int. Conference on Pattern Recognition, vol. 1, pp. 566–568 (1994)Google Scholar
  5. 5.
    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
  6. 6.
    Goldfarb, L., Gay, D., Golubitsky, O., Korkin, D.: What is a structural representation? second version. Technical Report TR04-165, University of New Brunswick, Fredericton, Canada (2004)Google Scholar
  7. 7.
    Duin, R.P.W.: Structural class representation and pattern recognition by ets; a commentary. Technical report, Delft University of Technology, Pattern Recognition Laboratory (2006)Google Scholar
  8. 8.
    Pękalska, 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
  9. 9.
    Duin, R.P.W., Pekalska, E., Harol, A., Lee, W.J., Bunke, H.: On euclidean corrections for non-euclidean dissimilarities. In: Structural, Syntactic, and Statistical Pattern Recognition, pp. 551–561 (2008)Google Scholar
  10. 10.
    Jacobs, D., Weinshall, D., Gdalyahu, Y.: Classification with Non-Metric Distances: Image Retrieval and Class Representation. IEEE TPAMI 22(6), 583–600 (2000)Google Scholar
  11. 11.
    Duin, R., de Ridder, D., Tax, D.: Experiments with object based discriminant functions; a featureless approach to pattern recognition. Pattern Recognition Letters 18(11-13), 1159–1166 (1997)CrossRefGoogle Scholar
  12. 12.
    Graepel, T., Herbrich, R., Bollmann-Sdorra, P., Obermayer, K.: Classification on pairwise proximity data. In: Advances in Neural Information System Processing, vol. 11, pp. 438–444 (1999)Google Scholar
  13. 13.
    Pękalska, E., Paclík, P., Duin, R.: A Generalized Kernel Approach to Dissimilarity Based Classification. J. of Machine Learning Research 2(2), 175–211 (2002)MATHCrossRefGoogle Scholar
  14. 14.
    Bunke, H., Shearer, K.: A graph distance metric based on the maximal common subgraph. Pattern Recognition Letters 19(3-4), 255–259 (1998)MATHCrossRefGoogle Scholar
  15. 15.
    Bunke, H., Bühler, U.: Applications of approximate string matching to 2D shape recognition. Pattern recognition 26(12), 1797–1812 (1993)CrossRefGoogle Scholar
  16. 16.
    Jain, A.K., Zongker, D.E.: Representation and recognition of handwritten digits using deformable templates. IEEE Transactions on Pattern Analysis and Machine Intelligence 19(12), 1386–1391 (1997)CrossRefGoogle Scholar
  17. 17.
    Kondor, R.I., Jebara, T.: A kernel between sets of vectors. In: Fawcett, T., Mishra, N. (eds.) ICML, pp. 361–368. AAAI Press, Menlo Park (2003)Google Scholar
  18. 18.
    Graepel, T., Herbrich, R., Schölkopf, B., Smola, A., Bartlett, P., Müller, K.R., Obermayer, K., Williamson, R.: Classification on proximity data with LP-machines. In: ICANN, pp. 304–309 (1999)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Robert P. W. Duin
    • 1
  • Elżbieta Pękalska
    • 2
  1. 1.Faculty of Electrical Engineering, Mathematics and Computer SciencesDelft University of TechnologyThe Netherlands
  2. 2.School of Computer ScienceUniversity of ManchesterUnited Kingdom

Personalised recommendations