Selecting Structural Base Classifiers for Graph-Based Multiple Classifier Systems

  • Wan-Jui Lee
  • Robert P. W. Duin
  • Horst Bunke
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5997)


Selecting a set of good and diverse base classifiers is essential for building multiple classifier systems. However, almost all commonly used procedures for selecting such base classifiers cannot be directly applied to select structural base classifiers. The main reason is that structural data cannot be represented in a vector space.

For graph-based multiple classifier systems, only using subgraphs for building structural base classifiers has been considered so far. However, in theory, a full graph preserves more information than its subgraphs. Therefore, in this work, we propose a different procedure which can transform a labelled graph into a new set of unlabelled graphs and preserve all the linkages at the same time. By embedding the label information into edges, we can further ignore the labels. By assigning weights to the edges according to the labels of their linked nodes, the strengths of the connections are altered, but the topology of the graph as a whole is preserved.

Since it is very difficult to embed graphs into a vector space, graphs are usually classified based on pairwise graph distances. We adopt the dissimilarity representation and build the structural base classifiers based on labels in the dissimilarity space. By combining these structural base classifiers, we can solve the labelled graph classification problem with a multiple classifier system. The performance of using the subgraphs and full graphs to build multiple classifier systems is compared in a number of experiments.


Label Graph Label Information Connection Matrix Feature Subset Selection Full Graph 
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.
    Bunke, H., Irniger, C., Neuhaus, M.: Graph Matching - Challenges and Potential Solutions. In: Roli, F., Vitulano, S. (eds.) ICIAP 2005. LNCS, vol. 3617, pp. 1–10. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  2. 2.
    Bunke, H., Riesen, K.: Graph Classification Based on Dissimilarity Space Embedding. In: da Vitoria Lobo, N., Kasparis, T., Roli, F., Kwok, J.T., Georgiopoulos, M., Anagnostopoulos, G.C., Loog, M. (eds.) SSSPR 2008. LNCS, vol. 5342, pp. 996–1007. Springer, Heidelberg (2008)CrossRefGoogle Scholar
  3. 3.
    Dijkstra, E.W.: A Note on Two Problems in Connexion with Graphs. Numerische Mathematik 1, 269–271 (1959)MATHCrossRefMathSciNetGoogle Scholar
  4. 4.
    Duin, R.P.W., Juszczak, P., Paclik, P., Pȩkalska, E., de Ridder, D., Tax, D.M.J.: PRTOOLS 2004, A Matlab Toolbox for Pattern Recognition. Delft University of Technology, ICT Group, The Netherlands (2004), Google Scholar
  5. 5.
    Ho, T.K.: The Random Subspace Method for Constructing Decision Forests. IEEE Trans. Pattern Analysis and Machine Intelligence 20(8), 832–844 (1998)CrossRefGoogle Scholar
  6. 6.
    Kuncheva, L.I.: Combining Pattern Classifiers. In: Methods and Algorithms. Wiley, Chichester (2004)Google Scholar
  7. 7.
    Lee, W.J., Duin, R.P.W.: An Inexact Graph Comparison Approach in Joint Eigenspace. In: da Vitoria Lobo, N., Kasparis, T., Roli, F., Kwok, J.T., Georgiopoulos, M., Anagnostopoulos, G.C., Loog, M. (eds.) SSSPR 2008. LNCS, vol. 5342, pp. 35–44. Springer, Heidelberg (2008)CrossRefGoogle Scholar
  8. 8.
    Lee, W.J., Duin, R.P.W.: A Labelled Graph Based Multiple Classifier System. In: Benediktsson, J.A., Kittler, J., Roli, F. (eds.) MCS 2009. LNCS, vol. 5519, pp. 201–210. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  9. 9.
    Neuhaus, M., Bunke, H.: Edit Distance-Based Kernel Functions for Structural Pattern Classification. Pattern Recognition 39, 1852–1863 (2006)MATHCrossRefGoogle Scholar
  10. 10.
    Pȩkalska, E., Duin, R.P.W.: The Dissimilarity Representation for Pattern Recognition. In: Fundations and Applications. World Scientific, Singapore (2005)Google Scholar
  11. 11.
    Qiu, H.J., Hancock, E.R.: Spectral Simplication of Graphs. In: Pajdla, T., Matas, J(G.) (eds.) ECCV 2004. LNCS, vol. 3024, pp. 114–126. Springer, Heidelberg (2004)Google Scholar
  12. 12.
    Riesen, K., Bunke, H.: Classifier Ensembles for Vector Space Embedding of Graphs. In: Haindl, M., Kittler, J., Roli, F. (eds.) MCS 2007. LNCS, vol. 4472, pp. 220–230. Springer, Heidelberg (2007)CrossRefGoogle Scholar
  13. 13.
    Riesen, K., Bunke, H.: IAM Graph Database Repository for Graph Based Pattern Recognition and Machine Learning. In: da Vitoria Lobo, N., Kasparis, T., Roli, F., Kwok, J.T., Georgiopoulos, M., Anagnostopoulos, G.C., Loog, M. (eds.) SSSPR 2008. LNCS, vol. 5342, pp. 287–297. Springer, Heidelberg (2008)CrossRefGoogle Scholar
  14. 14.
    Schenker, A., Bunke, H., Last, M., Kandel, A.: Building Graph-Based Classifier Ensembles by Random Node Selection. In: Roli, F., Kittler, J., Windeatt, T. (eds.) MCS 2004. LNCS, vol. 3077, pp. 214–222. Springer, Heidelberg (2004)Google Scholar
  15. 15.
    Skurichina, M., Kuncheva, L.I., Duin, R.P.W.: Bagging and Boosting for the Nearest Mean Classifier: Effects of Sample Sizes on Diversity an Accuracy. In: Roli, F., Kittler, J. (eds.) MCS 2002. LNCS, vol. 2364, pp. 62–71. Springer, Heidelberg (2002)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2010

Authors and Affiliations

  • Wan-Jui Lee
    • 1
  • Robert P. W. Duin
    • 1
  • Horst Bunke
    • 2
  1. 1.Pattern Recognition LaboratoryDelft University of TechnologyThe Netherlands
  2. 2.Institute of Computer Science and Applied MathematicsUniversity of BernSwitzerland

Personalised recommendations