Neural Processing Letters

, Volume 8, Issue 1, pp 1–8 | Cite as

Classification by Density Intersection

  • Yoram Baram
Article
  • 27 Downloads

Abstract

A classification method based on the intersection surface between two parameterized densities is proposed. The densities are obtained from class-labeled data by maximizing the mutual information across a system of integrated Gaussians, but, in practice, only the intersection surface needs to be estimated. The application of the proposed technique is demonstrated by predicting stock behavior.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    T.W. Anderson, An Introduction to Multivariate Statistical Analysis, John Wiley, New York, 1958.Google Scholar
  2. 2.
    A.C. Andrews, Mathematical Techniques in Pattern Recognition, Wiley-Interscience, New York, 1972.Google Scholar
  3. 3.
    Z. Roth and Y. Baram, “Multi-dimensional density shaping by sigmoids,” IEEE Trans. on Neural Networks, Vol. 7, No. 5, pp. 1291-1298, September 1996.Google Scholar
  4. 4.
    R. Battiti, “Using mutual information for selecting features in unsupervised neural net learning,” IEEE Trans. on Neural Networks, Vol. 5, No. 4, pp. 537-550, July 1994.Google Scholar
  5. 5.
    A.J. Bell and T.J. Sejnowski, “An information maximization approach to blind separation and blind deconvolution,” Neural Computation, Vol. 7, No. 6, pp. 1129-1159, 1995.Google Scholar
  6. 6.
    M. Bichsel and P. Seiz, “Minimum class entropy: a maximum entropy approach to layered networks,” Neural Networks, 2: 133-141, 1989.Google Scholar
  7. 7.
    C.M. Bishop, Neural Networks for Pattern Recognition, Clarendon Press, Oxford, 1995.Google Scholar
  8. 8.
    K. Fukunaga, Introduction to Statistical Pattern Recognition, Academic Press, San Diego, 1990.Google Scholar
  9. 9.
    P. G. Hoel, Introduction to Mathematical Statistics, Wiley, New York, 1984.Google Scholar
  10. 10.
    A.R. Horn and C.H. Johnson, Matrix Analysis, Cambridge University Press, 1985.Google Scholar
  11. 11.
    R. Linsker, “How to generate ordered maps by maximizing the mutual information between input and output signals,” Neural Computation, Vol. 1, No. 3, pp. 402-411, 1989.Google Scholar
  12. 12.
    G.H. Golub and C.F. Van Loan, Matrix Computation, The Johns Hopkins University Press, Baltimore, MD, 1983.Google Scholar
  13. 13.
    A. Papoulis, Probability, Random Variables and Stochastic Processes, McGraw-Hill, Princeton, 1984.Google Scholar

Copyright information

© Kluwer Academic Publishers 1998

Authors and Affiliations

  • Yoram Baram
    • 1
  1. 1.Computer Science Department, TechnionIsrael Institute of TechnologyHaifaIsrael

Personalised recommendations