A Family of Novel Graph Kernels for Structural Pattern Recognition

  • Horst Bunke
  • Kaspar Riesen
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4756)

Abstract

Recently, an emerging trend of representing objects by graphs can be observed. As a matter of fact, graphs offer a versatile alternative to feature vectors in pattern recognition, machine learning and data mining. However, the space of graphs contains almost no mathematical structure, and consequently, there is a lack of suitable methods for graph classification. Graph kernels, a novel class of algorithms for pattern analysis, offer an elegant solution to this problem. Graph kernels aim at bridging the gap between statistical and symbolic object representations. In the present paper we propose a general approach to transforming graphs into n-dimensional real vector spaces by means of graph edit distance. As a matter of fact, this approach results in a novel family of graph kernels making a wide range of kernel machines applicable for graphs. With several experimental results we prove the robustness and flexibility of our new method and show that our approach outperforms a standard graph classification method on several graph data sets of diverse nature.

Keywords

Feature Vector Linear Discriminant Analysis Edit Distance Edit Operation Graph Embedding 
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.
    Duda, R., Hart, P., Stork, D.: Pattern Classification, 2nd edn. Wiley-Interscience, Chichester (2000)Google Scholar
  2. 2.
    Schölkopf, B., Smola, A.: Learning with Kernels. MIT Press, Cambridge (2002)Google Scholar
  3. 3.
    Shawe-Taylor, J., Cristianini, N.: Kernel Methods for Pattern Analysis. Cambridge University Press, Cambridge (2004)Google Scholar
  4. 4.
    Vapnik, V., Chervonenkis, A.: On the uniform convergence of relative frequencies of events to their probabilities. Theory of Probability and its Applications 16(2), 264–280 (1971)CrossRefMathSciNetGoogle Scholar
  5. 5.
    Vapnik, V.: Statistical Learning Theory. John Wiley, Chichester (1998)MATHGoogle Scholar
  6. 6.
    Gärtner, T.: A survey of kernels for structured data. SIGKDD Explorations 5(1), 49–58 (2003)CrossRefGoogle Scholar
  7. 7.
    Riesen, K., Neuhaus, M., Bunke, H.: Graph embedding in vector spaces by means of prototype selection. In: Accepted for the 6th Int. Workshop on Graph-Based Representations in Pattern Recognition (2007)Google Scholar
  8. 8.
    Riesen, K., Kilchherr, V., Bunke, H.: Reducing the dimensionality of vector space embeddings of graphs. In: Accepted for the International Conference on Machine Learning and Data Mining MLDM (2007)Google Scholar
  9. 9.
    Riesen, K., Bunke, H.: Classifier ensembles for vector space embedding of graphs. In: Accepted for the 7th Int. Workshop on Multiple Classifier Systems (2007)Google Scholar
  10. 10.
    Gärtner, T., Flach, P., Wrobel, S.: On graph kernels: Hardness results and efficient alternatives. In: Schölkopf, B., Warmuth, M., eds.: Proc. 16th Annual Conf. on Learning Theory, pp. 129–143 (2003)Google Scholar
  11. 11.
    Borgwardt, K., Ong, C., Schönauer, S., Vishwanathan, S., Smola, A., Kriegel, H.P.: Protein function prediction via graph kernels. Bioinformatics 21(1), 47–56 (2005)CrossRefGoogle Scholar
  12. 12.
    Mahé, P., Ueda, N., Akutsu, T.: Graph kernels for molecular structures – activity relationship analysis with support vector machines. Journal of Chemical Information and Modeling 45(4), 939–951 (2005)CrossRefGoogle Scholar
  13. 13.
    Lafferty, J., Lebanon, G.: Diffusion kernels on statistical manifolds. Journal of Machine Learning Research 6, 129–163 (2005)MathSciNetGoogle Scholar
  14. 14.
    Haussler, D.: Convolution kernels on discrete structures. Technical Report UCSC-CRL-99-10, University of California, Santa Cruz (1999)Google Scholar
  15. 15.
    Watkins, C.: Kernels from matching operations. Technical Report CSD-TR-98-07, Royal Holloway College (1999)Google Scholar
  16. 16.
    Jain, B., Geibel, P., Wysotzki, F.: SVM learning with the Schur-Hadamard inner product for graphs. Neurocomputing 64, 93–105 (2005)CrossRefGoogle Scholar
  17. 17.
    Ramon, J., Gärtner, T.: Expressivity versus efficiency of graph kernels. In: Proc. First International Workshop on Mining Graphs, Trees and Sequences, 65–74 (2003)Google Scholar
  18. 18.
    Horvath, T., Gärtner, T., Wrobel, S.: Cyclic pattern kernels for predictive graph mining. In: Proc. International Conference on Knowledge Discovery and Data Mining, pp. 65–74. ACM Press, New York (2004)Google Scholar
  19. 19.
    Neuhaus, M.: Bridging the Gap Between Graph Edit Distance and Kernel Machines. World Scientific (to appear, 2007)Google Scholar
  20. 20.
    Bunke, H., Allermann, G.: Inexact graph matching for structural pattern recognition. Pattern Recognition Letters 1, 245–253 (1983)CrossRefGoogle Scholar
  21. 21.
    Sanfeliu, A., Fu, K.: A distance measure between attributed relational graphs for pattern recognition. IEEE Transactions on Systems, Man, and Cybernetics (Part B) 13(3), 353–363 (1983)MATHGoogle Scholar
  22. 22.
    Riesen, K., Neuhaus, M., Bunke, H.: Bipartite graph matching for computing the edit distance of graphs. In: Accepted for the 6th Int. Workshop on Graph-Based Representations in Pattern Recognition (2007)Google Scholar
  23. 23.
    Neuhaus, M., Riesen, K., Bunke, H.: Fast suboptimal algorithms for the computation of graph edit distance. In: Ribeiro, C.C., Martins, S.L. (eds.) WEA 2004. LNCS, vol. 3059, pp. 163–172. Springer, Heidelberg (2004)Google Scholar
  24. 24.
    Wilson, R., Hancock, E., Luo, B.: Pattern vectors from algebraic graph theory. IEEE Trans. on Pattern Analysis ans Machine Intelligence 27(7), 1112–1124 (2005)CrossRefGoogle Scholar
  25. 25.
    Wilson, R., Hancock, E.: Levenshtein distance for graph spectral features. In: Proc. 17th Int. Conf. on Pattern Recognition, vol. 2, pp. 489–492 (2004)Google Scholar
  26. 26.
    Luo, B., Wilson, R., Hancock, E.: Spectral embedding of graphs. Pattern Recognition 36(10), 2213–2223 (2003)MATHCrossRefGoogle Scholar
  27. 27.
    Duin, R., Pekalska, E.: The Dissimilarity Representations for Pattern Recognition: Foundations and Applications. World Scientific (2005)Google Scholar
  28. 28.
    Riesen, K., Bunke, H.: A graph kernel based on vector space embedding (Submitted)Google Scholar
  29. 29.
    Bishop, C.: Neural Networks for Pattern Recognition. Oxford University Press, Oxford (1996)MATHGoogle Scholar
  30. 30.
    Kuncheva, L.: Combining Pattern Classifiers: Methods and Algorithms. John Wiley, Chichester (2004)MATHGoogle Scholar
  31. 31.
    Breiman, L.: Bagging predictors. Machine Learning 24, 123–140 (1996)MATHMathSciNetGoogle Scholar
  32. 32.
    Ho, T.: The random subspace method for constructing decision forests. IEEE Trans. on Pattern Analysis ans Machine Intelligence 20(8), 832–844 (1998)CrossRefGoogle Scholar
  33. 33.
    Freund, Y., Shapire, R.: A decision theoretic generalization of online learning and application to boosting. Journal of Computer and Systems Sciences 55, 119–139 (1997)MATHCrossRefGoogle Scholar
  34. 34.
    Pudil, P., Novovicova, J., Kittler, J.: Floating search methods in feature-selection. Pattern Recognition Letters 15(11), 1119–1125 (1994)CrossRefGoogle Scholar
  35. 35.
    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)Google Scholar
  36. 36.
    Watson, C., Wilson, C.: NIST special database 4, fingerprint database. National Institute of Standards and Technology (1992)Google Scholar
  37. 37.
    Neuhaus, M., Bunke, H.: A graph matching based approach to fingerprint classification using directional variance. In: Kanade, T., Jain, A., Ratha, N.K. (eds.) AVBPA 2005. LNCS, vol. 3546, pp. 191–200. Springer, Heidelberg (2005)Google Scholar
  38. 38.
    DTP, D.T.P.: Aids antiviral screen (2004), http://dtp.nci.nih.gov/docs/aids/aids_data.html

Copyright information

© Springer-Verlag Berlin Heidelberg 2007

Authors and Affiliations

  • Horst Bunke
    • 1
  • Kaspar Riesen
    • 1
  1. 1.Institute of Computer Science and Applied Mathematics, University of Bern, Neubrückstrasse 10, CH-3012 BernSwitzerland

Personalised recommendations