Advertisement

On the Training Patterns Pruning for Optimum-Path Forest

  • João P. Papa
  • Alexandre X. Falcão
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5716)

Abstract

The Optimum-Path Forest (OPF) classifier is a novel graph-based supervised pattern recognition technique that has been demonstrated to be superior to Artificial Neural Networks and similar to Support Vector Machines, but much faster. The OPF classifier reduces the problem of pattern recognition to a computation of an optimum-path forest in the feature space induced by a graph, creating discrete optimal partitions, which are optimum-path trees rooted by prototypes, i.e., key samples that will compete among themselves trying to conquer the remaining samples. Some applications, such that medical specialist systems for image-based diseases identification, need to be constantly re-trained with new instances (diagnostics) to achieve a better generalization of the problem, which requires large storage devices, due to the high number of generated data (millions of voxels). In that way, we present here a pruning algorithm for the OPF classifier that learns the most irrelevant samples and eliminate them from the training set, without compromising the classifier’s accuracy.

Keywords

Optimum Path Oropharyngeal Dysphagia Pruning Algorithm Laryngeal Pathology COREL Dataset 
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.
    Papa, J.P., Falcão, A.X., Suzuki, C.T.N., Mascarenhas, N.D.A.: A discrete approach for supervised pattern recognition. In: Brimkov, V.E., Barneva, R.P., Hauptman, H.A. (eds.) IWCIA 2008. LNCS, vol. 4958, pp. 136–147. Springer, Heidelberg (2008)CrossRefGoogle Scholar
  2. 2.
    Haykin, S.: Neural networks: a comprehensive foundation. Prentice Hall, Englewood Cliffs (1994)zbMATHGoogle Scholar
  3. 3.
    Boser, B.E., Guyon, I.M., Vapnik, V.N.: A training algorithm for optimal margin classifiers. In: Proceedings of the 5th Workshop on Computational Learning Theory, pp. 144–152. ACM Press, New York (1992)Google Scholar
  4. 4.
    Papa, J.P., Spadotto, A.A., Falcão, A.X., Pereira, J.C.: Optimum path forest classifier applied to laryngeal pathology detection. In: Proc. of the 15th Intl. Conf. on Systems, Signals, and Image Processing, vol. 1, pp. 249–252. IEEE, Los Alamitos (2008)Google Scholar
  5. 5.
    Spadotto, A.A., Pereira, J.C., Guido, R.C., Papa, J.P., Falcão, A.X., Gatto, A.R., Cola, P.C., Schelp, A.O.: Oropharyngeal dysphagia identification using wavelets and optimum path forest. In: Proc. of the 3rd IEEE Intl. Symp. on Communications, Control and Signal Processing, pp. 735–740 (2008)Google Scholar
  6. 6.
    Montoya-Zegarra, J.A., Papa, J.P., Leite, N.J., Torres, R.S., Falcão, A.X.: Learning how to extract rotation-invariant and scale-invariant features from texture images. EURASIP Journal on Advances in Signal Processing 2008, 1–16 (2008)CrossRefzbMATHGoogle Scholar
  7. 7.
    Freitas, G.M., Ávila, A.M.H., Pinto, H.S., Papa, J.P., Falcão, A.X.: Optimum-path forest-based models for rainfall estimation. In: Proceedings of the 14th IEEE Symposium on Computers and Communications (submitted, 2009)Google Scholar
  8. 8.
    Kuncheva, L.I.: Combining Pattern Classifiers: Methods and Algorithms. Wiley-Interscience, Hoboken (2004)CrossRefzbMATHGoogle Scholar
  9. 9.
    Wang, Y.P., Pavlidis, T.: Optimal correspondence of string subsequences. IEEE Transactions on Pattern Analysis and Machine Intelligence 12(11), 1080–1087 (1990)CrossRefGoogle Scholar
  10. 10.
    Arica, N., Vural, F.T.Y.: BAS: A Perceptual Shape Descriptor Based on the Beam Angle Statistics. Pattern Recognition Letters 24(9-10), 1627–1639 (2003)CrossRefzbMATHGoogle Scholar
  11. 11.
    Stehling, R.O., Nascimento, M.A., Falcão, A.X.: A compact and efficient image retrieval approach based on border/interior pixel classification. In: Proceedings of the 11th International Conference on Information and Knowledge Management, pp. 102–109. ACM Press, New York (2002)Google Scholar
  12. 12.
    Falcão, A.X., Stolfi, J., Lotufo, R.A.: The image foresting transform: Theory, algorithms, and applications. IEEE Transactions on Pattern Analysis and Machine Intelligence 26(1), 19–29 (2004)CrossRefGoogle Scholar
  13. 13.
    Cormen, T., Leiserson, C., Rivest, R.: Introduction to Algorithms. MIT, Cambridge (1990)zbMATHGoogle Scholar
  14. 14.
    Rocha, A., Miranda, P.A.V., Falcão, A.X., Bergo, F.P.G.: Object delineation by κ-connected components. EURASIP Journal on Advances in Signal Processing 2008, 1–5 (2008)zbMATHGoogle Scholar
  15. 15.
    Chang, C.C., Lin, C.J.: LIBSVM: A Library for Support Vector Machines (2001), http://www.csie.ntu.edu.tw/~cjlin/libsvm
  16. 16.
    Nissen, S.: Implementation of a Fast Artificial Neural Network Library (FANN), Department of Computer Science University of Copenhagen, DIKU (2003), http://leenissen.dk/fann/
  17. 17.
    Papa, J.P., Suzuki, C.T.N., Falcão, A.X.: LibOPF: A library for the design of optimum-path forest classifiers (2008), Software version 1.0, http://www.ic.unicamp.br/~afalcao/LibOPF
  18. 18.
    MPEG-7. Mpeg-7: The generic multimedia content description standard, part 1. IEEE MultiMedia 09(2), 78–87 (2002)Google Scholar
  19. 19.
    Corel Corporation. Corel stock photo images, http://www.corel.com
  20. 20.
    Kuncheva, L.: Artificial data. School of Informatics, University of Wales, Bangor (1996), http://www.informatics.bangor.ac.uk/~mas00a
  21. 21.
    Persoon, E., Fu, K.: Shape Discrimination Using Fourier Descriptors. IEEE Transanctions on Systems, Man, and Cybernetics 7(3), 170–178 (1977)MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2009

Authors and Affiliations

  • João P. Papa
    • 1
  • Alexandre X. Falcão
    • 1
  1. 1.Institute of Computing, Visual Informatics LaboratoryUniversity of CampinasCampinasBrazil

Personalised recommendations