Some Experiments in Supervised Pattern Recognition with Incomplete Training Samples

  • Ricardo Barandela
  • Francesc J. Ferri
  • Tania Nájera
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 2396)


This paper presents some ideas about automatic procedures to implement a system with the capability of detecting patterns arising from classes not represented in the training sample. The procedure aims at incorporating automatically to the training sample the necessary information about the new class for correctly recognizing patterns from this class in future classification tasks. The Nearest Neighbor rule is employed as the central classifier and several techniques are added to cope with the peril of incorporating noisy data to the training sample. Experimental results with real data confirm the benefits of the proposed procedure.


Training Sample Outlier Detection Near Neighbor Training Pattern Generalize Edition 
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.


  1. 1.
    Dasarathy, B.V.: All you need to know about the neighbors. Proceedings International Conference on Cybernetics and Society, Denver (1979)Google Scholar
  2. 2.
    Dasarathy, B.V.: “There goes the neighborhood”: An Alien identification approach to recognition in partially exposed environments. Proceedings of the 5 th International Conference on Pattern Recognition (1980)Google Scholar
  3. 3.
    Wilson, D.L.: Asymptotic properties of Nearest Neighbor rules using edited data sets. IEEE Trans. Systems, Man and Cybernetics, SMC-2 (1972) 408–421CrossRefGoogle Scholar
  4. 4.
    Barandela, R.: La regla NN con muestras de entrenamiento no balanceadas. Investigación Operacional, X(1) (1990) 45–56Google Scholar
  5. 5.
    Sanchez, J.S., F. Pla and F. Ferri: Prototype selection for the Nearest Neighbor rule through Proximity Graphs. Pattern Recognition Letters, 18,6 (1997) 507–513CrossRefGoogle Scholar
  6. 6.
    Barandela, R.: Una metodología para el reconocimiento de patrones en tareas geólogo-geofísicas. Geofísica Internacional, 34(4) (1995) 399–405Google Scholar
  7. 7.
    Gasca, E. and R. Barandela: Influencia del preprocesamiento de la muestra de entrenamiento en el poder de generalización del Perceptron Multicapa. 6th Brazilian Symposium on Neural Networks. Río de Janeiro (2000)Google Scholar
  8. 8.
    Barandela, R. and E. Gasca: Decontamination of training samples for supervised pattern recognition methods. In: Advances in Pattern Recognition, Lecture Notes in Computer Science 1876, Springer Verlag (2000) 621–630CrossRefGoogle Scholar
  9. 9.
    Koplowitz, J. and T.A. Brown: On the relation of performance to editing in nearest neighbor rules. Proc. 4th Int. Joint Conf. on Pattern Recognition, Japan (1978)Google Scholar
  10. 10.
    Dasarathy, B.V.: Adaptive decision systems with extended learning for deployment in partially exposed environments. Optical Engineering, 34,5, (1995) 1269–1280CrossRefGoogle Scholar
  11. 11.
    Tax, D.M.J. and R.P.W. Duin: Outlier Detection using Classifier Instability. In: A. Amin et al. (eds.), Advances in Pattern Recognition, Lecture Notes in Computer Science, vol 1451, Springer, Berlin (1998)CrossRefGoogle Scholar
  12. 12.
    Merz, C.J. and P.M. Murphy: UCI Repository of machine learning databases. University of California at Irvine. (1998)Google Scholar
  13. 13.
    Wilson, D.R. and T.R. Martinez: Reduction Techniques for Instance-Based Learning Algorithms. Machine Learning, 38,3 (2000) 257–286zbMATHCrossRefGoogle Scholar
  14. 14.
    Duda, R.O. and P.E. Hart: Pattern Classification and Scene Analysis. Wiley, New York (1973)zbMATHGoogle Scholar
  15. 15.
    Barandela, R. and M. Juarez: Ongoing learning for Supervised Pattern Recognition. Proceedings of the XIV SIBGRAPI, Brazil, IEEE Press (2001) 51–58CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2002

Authors and Affiliations

  • Ricardo Barandela
    • 1
  • Francesc J. Ferri
    • 2
  • Tania Nájera
    • 1
  1. 1.Lab for Pattern RecognitionInstituto Tecnológico de Toluca, MéxicoMetepecEstado de México
  2. 2.Dept. d’InformaticaUniversitat de ValènciaBurjassot, ValènciaSpain

Personalised recommendations