Editing prototypes in the finite sample size case using alternative neighborhoods

  • F. J. Ferri
  • J. S. Sánchez
  • F. Pla
Statistical Classification Techniques
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1451)


The recently introduced concept of Nearest Centroid Neighborhood is applied to discard outliers and prototypes 111 class overlapping regions in order to improve the performance of the Nearest Neighbor rule through an editing procedure, This approach is related to graph based editing algorithms which also define alternative neighborhoods in terms of geornetric relations, Classical editing algorithms are compared to these alternative editing schemes using several synthetic and real data problems. The empirical results show that, the proposed editing algorithm constitutes a good trade-off among performance and computational burden.


Near Neighbor Delaunay Triangulation Neighbor Rule Pattern Recognition Letter Prototype Selection 
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.
    B.B. Chandhuri. A new definition of neighbourhood of a point in multi-dimensional space. Pattern Recognition Letters, 17:11–17, 1996.CrossRefGoogle Scholar
  2. 2.
    T. M. Cover and P. E. Hart. Nearest neighbor pattern classification, IEEE Transactions on Information Theory, 13:21–27, 1967.CrossRefGoogle Scholar
  3. 3.
    B. V. Dasarathy and B. V. Sheela. Visiting nearest neighbors. In Proc. Int. Conf. on Cybernetics and society, pages 630–635, 1977.Google Scholar
  4. 4.
    P. A. Devijver and J. Kittler. Pattern Recognition. A Statistical Approach. Prentice Hall, 1982.Google Scholar
  5. 5.
    F. Ferri and E. Vidal. Small sample size effects in the use of editing techniques. In Proc. of 11th International Conference of Pattern Recognition, pages 607–610, The Hague, THE NETHERLANDS, September 1992.Google Scholar
  6. 6.
    B.K. Bhattacharya G.T. Toussaint and R.S. Poulsen. The application of voronoi diagrams to nonparametric decision rules. In L. Billard, editor, Computer Science and Statistics: The Interface, Elsevier Science, North-Holland, 1985.Google Scholar
  7. 7.
    L. Kuncheva. Editing for the k-nearest neighbors rule by a genetic algorithm, Pattern Recognition Letters, 16(8):809–814, 1995.CrossRefGoogle Scholar
  8. 8.
    A.E. Lucas and J, Kittler. A comparative study of the kohonen and multiedit neural net learning algorithms, In Proc. 1st IEE Int. Conf.on Artificial Neural Networks, pages 7–11, 1991.Google Scholar
  9. 9.
    J.E.S. Macleod, A. Luck, and D.M. Titterington. A re-examination of the distance-weighted k-nearest-neighbor classification rule. IEEE Transactions on Systems Man and Cybernetics, 17(4):689–696, 1987.Google Scholar
  10. 10.
    J.S. Sánchez, F. Pla, and F.J. Ferri. Prototype selection for the nearest neighbour rule through proximity graphs. Pattern Recognition Letters, 18(7):507–513, 1997.CrossRefGoogle Scholar
  11. 11.
    J.S. Sánchez, F. Pla, and F.J. Ferri. On the use of neighbourhood-based non-parametric classifiers. Pattern Recognition Letters, (in press), 1998.Google Scholar
  12. 12.
    R.D. Short and K. Fukunaga. The optimal distance measure for nearest neighbor classification. IEEE Transactions on Information Theory, 27(5):622–627, 1981.CrossRefGoogle Scholar
  13. 13.
    J. Voisin and P. A. Devijver. An application of the multiedit-condensing technique to the reference selection problem in a print recognition system. Pattern Recognition, 20(5):465–474, 1987.CrossRefGoogle Scholar
  14. 14.
    D. L. Wilson. Asymptotic properties of nearest neighbor rules using edited data. IEEE Transactions on Systems Man and Cybernetics, 2(3):408–421, 1972.Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1998

Authors and Affiliations

  • F. J. Ferri
    • 1
  • J. S. Sánchez
    • 2
  • F. Pla
    • 2
  1. 1.Institut de RobòticaUniversitat de ValènciaBurjassotSpain
  2. 2.Departament d'InfornàticaUniversitat Jaume ICastellóSpain

Personalised recommendations