Principal Components and the Accuracy of Machine Learning

  • Zbigniew Duszak
  • Waldemar W. Koczkodaj
Conference paper


Induction algorithms could be positioned somewhere between witchcraft and science since extracting rules (the essence of data semantics) from an incomplete set of examples is a devil’s task. On the other hand there always has been some romanticism in the interpretation of eigenvalues. Authors of this paper try to show that the abovfe mentioned notions work together quite well. Quite surprising results have been achieved. Not only the number of decision variables can be reduced but, in most tested cases, the prediction accuracy improves (despite the reduced number of variables). The results could be implemented as a preprocessor for most of the induction algorithms employed for an automatic rule generation (one of the most important components of any modern expert system).


Expert System Induction Algorithm Decision Tree Induction Multivariate Discriminant Analysis Knowledge Representation System 
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.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. [1]
    de Mantaras, R.L. — A Distance-Based Attribute Selection Measure for Decision Tree Induction, Machine Learning, 6, 81–92 (1991).CrossRefGoogle Scholar
  2. [2]
    Fukunaga K. — Introduction to Statistical Pattern Recognition, Academic Press, 1972.Google Scholar
  3. [3]
    Grzymala-Busse, J. — Managing Uncertainty in Expert Systems, Kluwer Academic Publ., 1991.Google Scholar
  4. [4]
    Reynolds, R. Koczkodaj, W.W. (eds) — Special Issue on Machine Learning and Induction Algorithms, International Journal of Software Engineering and Knowledge Engineering, Vol. 1, No. 4 (December 1991) pp. 327–480.CrossRefGoogle Scholar

Copyright information

© Springer-Verlag/Wien 1992

Authors and Affiliations

  • Zbigniew Duszak
    • 1
  • Waldemar W. Koczkodaj
    • 2
  1. 1.Elliot Lake Research Field StationElliot LakeCanada
  2. 2.Department of Mathematics and Computer ScienceLaurentian UniversitySudburyCanada

Personalised recommendations