Improving RBF networks by the feature selection approach EUBAFES

  • M. Scherf
  • W. Brauer
Part III: Learning: Theory and Algorithms
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1327)


The curse of dimensionality is one of the severest problems concerning the application of RBF networks. The number of RBF nodes and therefore the number of training examples needed grows exponentially with the intrinsic dimensionality of the input space. One way to address this problem is the application of feature selection as a data pre processing step.

In this paper we propose a two-step approach for the determination of an optimal feature subset: First, all possible feature-subsets are reduced to those with best discrimination properties by the application of the fast and robust filter technique EUBAFES. Secondly we use a wrapper approach to judge, which of the pre-selected feature subsets leads to RBF networks with least complexity and best classification accuracy. Experiments are undertaken to show the improvement for RBF networks by our feature selection approach.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Berthold MR., Diamond J., Boosting the Performance of RBF Networks with Dynamic Decay Adjustment, Advances in Neural Information Processing Systems, vol.7 (1995)Google Scholar
  2. 2.
    John G., Kohavi R., Pfleger K., Irrelevant Features and the Subset Selection Problem, Machine Learning: Proceedings of the Eleventh International Conference, (1993) 121–129 William W. Cohne and Haym HirshGoogle Scholar
  3. 3.
    Kira K., Rendell L.: A practical approach to feature selection. Proceedings of the International Conference on Machine Learning, Aberdeen, (1992) 249–256,Sleeman D., Edwards P. Morgan KaufmannGoogle Scholar
  4. 4.
    Kononenko I., Estimation attributes: Analysis and extensions of RELIEF in Proceedings of the European Conference on Machine Learning, Catana, Italy (1994) 171–182 Springer Verlag”Google Scholar
  5. 5.
    Moody J. and Darken C.J., Fast learning in networks with locally-tuned processing units Neural Computation, 1, (1989) 281–294Google Scholar
  6. 6.
    Scherf M., Brauer W., Feature Selection by Means of a Feature Weighting Approach. Technical Report No. FKI-221-97, Forschungsberichte künstliche Intelligenz, Institut für Informatik, Technische Universität München (1997)Google Scholar
  7. 7.
    Wettschereck D., Aha D.W., Mohori T., A Review and Empirical Evaluation of Feature Weighting Methods for a Class of Lazy Learning Algorithms. Artificial Intelligence Review (to appear) (1997)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1997

Authors and Affiliations

  • M. Scherf
    • 1
  • W. Brauer
    • 2
  1. 1.GSFNational Research Center for Environment and Health, medis InstituteNeuherberg
  2. 2.Institut für InformatikTechnische Universität MünchenMünchen

Personalised recommendations