Advertisement

On Prediction Mechanisms in Fast Branch & Bound Algorithms

  • Petr Somol
  • Pavel Pudil
  • Jiří Grim
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3138)

Abstract

The idea of using the Branch & Bound search for optimal feature selection has been recently refined by introducing additional predicting heuristics that is able to considerably accelerate the search process while keeping the optimality of results unaffected. The heuristics is used most extensively in the so-called Fast Branch & Bound algorithm, where it replaces many slow criterion function computations by means of fast predictions. In this paper we investigate alternative prediction mechanisms. The alternatives are shown potentially useful for simplification and speed-up of the algorithm. We demonstrate the robustness of the prediction mechanism concept on real data experiments.

Keywords

subset search feature selection search tree optimal search subset selection dimensionality reduction 

References

  1. 1.
    Devijver, P.A., Kittler, J.: Pattern Recognition: A Statistical Approach. Prentice-Hall, Englewood Cliffs (1982)zbMATHGoogle Scholar
  2. 2.
    Fukunaga, K.: Introduction to Statistical Pattern Recognition, 2nd edn. Academic Press, Inc., London (1990)zbMATHGoogle Scholar
  3. 3.
    Hamamoto, Y., Uchimura, S., Matsuura, Y., Kanaoka, T., Tomita, S.: Evaluation of the branch and bound algorithm for feature selection. Pattern Recognition Letters 11(7), 453–456 (1990)zbMATHCrossRefGoogle Scholar
  4. 4.
    Kudo, M., Sklansky, J.: Comparison of algorithms that select features for pattern classifiers. Pattern Recognition 33(I), 25–41 (2000)CrossRefGoogle Scholar
  5. 5.
    Narendra, P.M., Fukunaga, K.: A branch and bound algorithm for feature subset selection. IEEE Transactions on Computers C-26, 917–922 (1977)CrossRefGoogle Scholar
  6. 6.
    Somol, P., Pudil, P., Ferri, F.J., Kittler, J.: Fast Branch & Bound Algorithm in Feature Selection. In: Proc. 4th World Multiconference on Systemics, Cybernetics and Informatics SCI 2000, Orlando, Florida, vol. VII, Part 1, pp. 646–651 (2000)Google Scholar
  7. 7.
    Somol, P., Pudil, P., Grim, J.: Branch & Bound Algorithm with Partial Prediction For Use with Recursive and Non-Recursive Criterion Forms. In: Singh, S., Murshed, N., Kropatsch, W.G. (eds.) ICAPR 2001. LNCS, vol. 2013, pp. 230–238. Springer, Heidelberg (2001)Google Scholar
  8. 8.
    Somol, P., Pudil, P., Kittler, J.: Fast Branch & Bound Algorithms in Feature Selection. To appear in IEEE Transactions on PAMI (July 2004)Google Scholar
  9. 9.
    Yu, B., Yuan, B.: A more efficient branch and bound algorithm for feature selection. Pattern Recognition 26, 883–889 (1993)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2004

Authors and Affiliations

  • Petr Somol
    • 1
    • 2
  • Pavel Pudil
    • 1
    • 2
  • Jiří Grim
    • 1
  1. 1.Dept. of Pattern Recognition, Institute of Information Theory and AutomationAcademy of Sciences of the Czech RepublicPrague
  2. 2.Faculty of Management of the Prague University of EconomicsCzech Republic

Personalised recommendations