Random walk kernels in conjunction with Support Vector Machines are powerful methods for error-tolerant graph matching. Because of their local definition, however, the applicability of random walk kernels strongly depends on the characteristics of the underlying graph representation. In this paper, we describe a simple extension to the standard random walk kernel based on graph edit distance. The idea is to include global matching information in the local similarity evaluation of random walks in graphs. The proposed extension allows us to improve the performance of the random walk kernel significantly. We present an experimental evaluation of our method on three difficult graph datasets.


Support Vector Machine Edit Distance Graph Match Edit Operation Graph Kernel 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Conte, D., Foggia, P., Sansone, C., Vento, M.: Thirty years of graph matching in pattern recognition. Int. Journal of Pattern Recognition and Artificial Intelligence 18, 265–298 (2004)CrossRefGoogle Scholar
  2. 2.
    Shawe-Taylor, J., Cristianini, N.: Kernel Methods for Pattern Analysis. Cambridge University Press, Cambridge (2004)Google Scholar
  3. 3.
    Watkins, C.: Dynamic alignment kernels. In: Smola, A., Bartlett, P., Schölkopf, B., Schuurmans, D. (eds.) Advances in Large Margin Classifiers, pp. 39–50. MIT Press, Cambridge (2000)Google Scholar
  4. 4.
    Haussler, D.: Convolution kernels on discrete structures. Technical Report UCSC-CRL-99-10, University of California, Santa Cruz (1999)Google Scholar
  5. 5.
    Jain, B., Geibel, P., Wysotzki, F.: SVM learning with the Schur-Hadamard inner product for graphs. Neurocomputing 64, 93–105 (2005)CrossRefGoogle Scholar
  6. 6.
    Gärtner, T., Flach, P., Wrobel, S.: On graph kernels: Hardness results and efficient alternatives. In: Schölkopf, B., Warmuth, M.K. (eds.) COLT/Kernel 2003. LNCS, vol. 2777, pp. 129–143. Springer, Heidelberg (2003)CrossRefGoogle Scholar
  7. 7.
    Borgwardt, K., Ong, C., Schönauer, S., Vishwanathan, S., Smola, A., Kriegel, H.P.: Protein function prediction via graph kernels. Bioinformatics 21, 47–56 (2005)CrossRefGoogle Scholar
  8. 8.
    Sanfeliu, A., Fu, K.: A distance measure between attributed relational graphs for pattern recognition. IEEE Transactions on Systems, Man, and Cybernetics (Part B) 13, 353–363 (1983)MATHGoogle Scholar
  9. 9.
    Bunke, H., Allermann, G.: Inexact graph matching for structural pattern recognition. Pattern Recognition Letters 1, 245–253 (1983)MATHCrossRefGoogle Scholar
  10. 10.
    Neuhaus, M., Bunke, H.: An error-tolerant approximate matching algorithm for attributed planar graphs and its application to fingerprint classification. In: Fred, A., Caelli, T.M., Duin, R.P.W., Campilho, A.C., de Ridder, D. (eds.) SSPR&SPR 2004. LNCS, vol. 3138, pp. 180–189. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  11. 11.
    Le Saux, B., Bunke, H.: Feature selection for graph-based image classifiers. In: Marques, J.S., Pérez de la Blanca, N., Pina, P. (eds.) IbPRIA 2005. LNCS, vol. 3523, pp. 147–154. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  12. 12.
    Ambauen, R., Fischer, S., Bunke, H.: Graph edit distance with node splitting and merging and its application to diatom identification. In: Hancock, E.R., Vento, M. (eds.) GbRPR 2003. LNCS, vol. 2726, pp. 95–106. Springer, Heidelberg (2003)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Michel Neuhaus
    • 1
  • Horst Bunke
    • 1
  1. 1.Institute of Computer Science and Applied MathematicsUniversity of BernBernSwitzerland

Personalised recommendations