Skip to main content
SpringerLink
Log in

A study of methods of classifier construction and updating

  • Published:
Automatic Control and Computer Sciences Aims and scope Submit manuscript

Abstract

This paper investigates the behavior of the inductive algorithm CART2 applied to constructing a classifier using a modern DBMS and solving problems related to classifier incremental updating. As testing objects we used four databases taken from different knowledge domains. The results of the experiments demonstrate good performance for the incrementally obtained classifier—i.e., for the Hypoteroid database (classification of thyroid body diseases), the identification error amounted to 1.95% (for 2870 samples) versus 1.8% for the classifier constructed using the full teaching selection.

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

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

References

  1. Maloof, M. and Michalski, R., Incremental Learning with Partial Instance Memory, Presentation on XIII International Symposium on Methodologies for Intelligent Systems, Lyon, France, 27 June, 2002.

  2. Maloof, M. and Michalski, R., Selecting examples for partial memory learning, Machine Learning, 2000, no. 41, pp. 27–52.

  3. Ferrer-Troyano, F., Aguilar-Ruiz, J.S., and Riquelme, J.C., Incremental Rule Learning based on Example Nearness from Numerical Data Streams, Proceedings of the SAC’05, Santa Fe, New Mexico, USA, March 13–17, 2005.

    Google Scholar 

  4. Shafer, J., Agrawal, R., and Mehta, M., SPRINT: a Scalable Parallel Classifier for Data Minings, Proceedings of the 22nd VLDB Conference, 1996.

  5. Bongard, M.M., Vaintsvaig, M.N., Guberman, Sh.A., Izvekova, M.L., and Smirnov, M.S., Use of Learning Program for Identification of Oil-Bearing Strata, Geologiya i geopfizika (Geology and Geophysics), 1966, no. 6, pp. 96–109.

  6. Breiman, L., Friedman, J.H., Olshen, R.A. and Stone, C.J., Classifications and regression trees, Belmont, CA: Wadswonth International, 1984.

    Google Scholar 

  7. Kornienko, Y. and Borisov, A., The CART methodology for production rules induction, 5th International Conference on Soft Computing MENDEL’99, Brno, June 9–12, 1999, pp. 362–366.

  8. Kornienko, Y. and Borisov, A., The CART2 inductive algorithm in comparison with standard machine learning methods, Proceedings of the International Scientific-Technical Workshop “Problems of Transfer Technology”, Ufa, September 30–October 1, 1999, pp. 154–161.

  9. Kornijenko, J., Dzenis, J. and Borisov, A., Application of inductive diagnostic rules to intracerebral extravasation analysis, Rīgas Tehniskās universitātes zinātniskie raksti. 5. sārija “Datorzinātne”, 20.sējums “Informācijas tehnologija un vadības zinātne”, R ga: RTU, 2004, p. 36–42.

  10. Internet: http://archive.ics.uci.edu/ml/.

Download references

Author information

Authors and Affiliations

  1. Riga Technical University, Kal’kyu St. 1, LV-1658, Riga, Latvia

    A. N. Borisov & Yu. V. Kornienko

Authors
  1. A. N. Borisov
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Yu. V. Kornienko
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to A. N. Borisov.

Additional information

Original Russian Text © A.N. Borisov, Yu.V. Kornienko, 2008, published in Avtomatika i Vychislitel’naya Tekhnika, 2008, No. 6, pp. 29–37.

About this article

Cite this article

Borisov, A.N., Kornienko, Y.V. A study of methods of classifier construction and updating. Aut. Conrol Comp. Sci. 42, 300–305 (2008). https://doi.org/10.3103/S0146411608060047

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.3103/S0146411608060047

Key words

Search

Navigation