Skip to main content
SpringerLink
Log in

Well-behaved evaluation functions for numerical attributes

  • Communications Session 2A Learning and Discovery Systems
  • Conference paper
  • First Online:
Foundations of Intelligent Systems (ISMIS 1997)

Part of the book series: Lecture Notes in Computer Science ((LNAI,volume 1325))

Included in the following conference series:

  • 87 Accesses

Abstract

The class of well-behaved evaluation functions simplifies and makes efficient the handling of numerical attributes; for them it suffices to concentrate on the boundary points in searching for the optimal partition. This holds always for binary partitions and also for multisplits if only the function is cumulative in addition to being well-behaved. A large portion of the most important attribute evaluation functions are well-behaved. This paper surveys the class of well-behaved functions. As a case study, we examine the properties of C4.5's attribute evaluation functions. Our empirical experiments show that a very simple cumulative rectification to the poor bias of information gain significantly outperforms gain ratio.

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Log in via an institution

Institutional subscriptions

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Auer P. Holte, R., Maass, W.: Theory and application of agnostic PAC-learning with small decision trees. In A. Prieditis, S. Russell (eds.), Proc. Twelfth Intl. Conf. on Machine Learning (21-29). Morgan Kaufmann, San Francisco, CA, 1995.

    Google Scholar 

  2. Breiman, L.: Some properties of splitting criteria. Mach. Learn. 24 (1996) 41–47.

    Google Scholar 

  3. Breiman, L., Friedman, J., Olshen, R., Stone, C.: Classification and Regression Trees. Wadsworth, Pacific Grove, CA, 1984.

    Google Scholar 

  4. Catlett, J.: On changing continuous attributes into ordered discrete attributes. In Y. Kodratoff (ed.), Proc. Fifth Europ. Working Session on Learning (164–178), Lecture Notes in Computer Science 482. Springer-Verlag, Berlin, 1991.

    Google Scholar 

  5. Elomaa, T., Rousu, J.: General and efficient multisplitting of numerical attributes. Report C-1996-82. Dept. of Computer Science, University of Helsinki. 1996, 25 pp.

    Google Scholar 

  6. Elomaa, T., Rousu, J.: On the well-behavedness of important attribute evaluation functions. In G. Grahne (ed.), Proc. Sixth Scand. Conf. on Artificial Intelligence (in press). IOS Press, Amsterdam, 1997.

    Google Scholar 

  7. Fayyad, U., Irani, K.: On the handling of continuous-valued attributes in decision tree generation. Mach. Learn. 8 (1992) 87–102.

    Google Scholar 

  8. Fayyad, U., Irani, K.: Multi-interval discretization of continuous-valued attributes for classification learning. In Proc. Thirteenth Intl. Joint Conf. on Artificial Intelligence (1022-1027). Morgan Kaufmann, San Mateo, CA, 1993.

    Google Scholar 

  9. Fulton, T., Kasif, S., Salzberg, S.: Efficient algorithms for finding multi-way splits for decision trees. In A. Prieditis, S. Russell (eds.), Proc. Twelfth Intl. Conf. on Machine Learning (244-251). Morgan Kaufmann, San Francisco, CA, 1995.

    Google Scholar 

  10. Kononenko, I.: On biases in estimating multi-valued attributes. In Proc. Fourteenth Intl. Joint Conf. on Artificial Intelligence (1034-1040). Morgan Kaufmann, San Francisco, CA, 1995.

    Google Scholar 

  11. López de Màntaras, R.: A distance-based attribute selection measure for decision tree induction. Mach. Learn. 6 (1991) 81–92.

    Google Scholar 

  12. Merz, C., Murphy, P.: UCI repository of machine learning databases (http://www.ics.uci.edu/ mlearn/MLRepository.html). 2 Dept. of Information and Computer Science, University of California at Irvine.

    Google Scholar 

  13. Quinlan, R.: Induction of decision trees. Mach. Learn. 1 (1986) 81–106.

    Google Scholar 

  14. Quinlan, R.: C4.5: Programs for Machine Learning. Morgan Kaufmann, San Mateo, CA, 1993.

    Google Scholar 

  15. Quinlan, R.: Improved use of continuous attributes in C4.5. J. Artif. Intell. Res. 4 (1996) 77–90.

    Google Scholar 

  16. Van de Merckt, T.: Decision trees in numerical attribute spaces. In Proc. Thirteenth Intl. Joint Conf. on Artificial Intelligence (1016–1021). Morgan Kaufmann, San Mateo, CA, 1993.

    Google Scholar 

  17. Wallace, C., Patrick, J.: Coding decision trees. Mach. Learn. 11 (1993) 7–22. *** DIRECT SUPPORT *** A0008166 00004

    Google Scholar 

Download references

Author information

Authors and Affiliations

  1. Institute for Systems, Informatics and Safety, Joint Research Centre European Commission, TP 270, I-21020, Ispra (VA), Italy

    Tapio Elomaa

  2. VTT Biotechnology ana Food Research, Tietotie 2, P. O. Box 1501, FIN-02044, VTT, Finland

    Juho Rousu

Authors
  1. Tapio Elomaa
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Juho Rousu
    View author publications

    You can also search for this author in PubMed Google Scholar

Editor information

Zbigniew W. Raś Andrzej Skowron

Rights and permissions

Reprints and permissions

Copyright information

© 1997 Springer-Verlag Berlin Heidelberg

About this paper

Cite this paper

Elomaa, T., Rousu, J. (1997). Well-behaved evaluation functions for numerical attributes. In: Raś, Z.W., Skowron, A. (eds) Foundations of Intelligent Systems. ISMIS 1997. Lecture Notes in Computer Science, vol 1325. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-63614-5_14

Download citation

  • DOI: https://doi.org/10.1007/3-540-63614-5_14

  • Published:

  • Publisher Name: Springer, Berlin, Heidelberg

  • Print ISBN: 978-3-540-63614-4

  • Online ISBN: 978-3-540-69612-4

  • eBook Packages: Springer Book Archive

Publish with us

Policies and ethics

Search

Navigation