Advertisement

Machine learning usefulness relies on accuracy and self-maintenance

  • Oscar Luaces
  • Jaime Alonso
  • Enrique A. de la Cal
  • José Ranilla
  • Antonio Bahamonde
3 Machine Learning Machine Learning Applications: Tools and Methods
Part of the Lecture Notes in Computer Science book series (LNCS, volume 1416)

Abstract

A new machine learning system, INNER, is presented in this paper. The system starts out from a collection of training examples; some of them are inflated generalizing their description so as to obtain a first draft of classification rules. An optimization stage, borrowed from our previous system, Fan, is then applied to return the final set of rules. The main goal of Inner, besides its high level of accuracy, is its ability for self-maintenance. To close the paper, we present a number of different experiments carried` out with INNER to illustrate how good the performance and stability of the system is.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

Bibliography

  1. [Aha, 90]
    AHA, D. W.: A Study of Instance-based Algorithms for Supervised Learning Tasks: Mathematical, Empirical, and Psychological Evaluations. Ph. D. Dissertation. University of California at Irvine, (1990).Google Scholar
  2. [Alonso, García, Bahamonde, 97]
    ALONSO GONZÁLEZ, J.; GARCIA ÁLVAREZ, A., and BAHAMONDE, A., Un sistema que adecúa las reglas aprendidas a las críticas recibidas sobre sus actuaciones. Proceedings of VII Conferencia de la Asociación Española para la Inteligencia Artificial, CAEPIA-97. Málaga, Spain, November, (1997).Google Scholar
  3. [Bahamonde, de la Cal, Ranilla, Alonso, 97]
    BAHAMONDE, A., DE LA CAL, ENRIQUE A., RANILLA, JOSE, ALONSO, JAIME: Self-organizing symbolic learned rules. In Biological and Artificial Computation: From Neuroscience to Technology, (Mira, MorenoDiaz, Cabestany Eds.), pp. 536–545. LNCS No 1240, Springer-Verlag. Berlin (1997).Google Scholar
  4. [Fisher, 36]
    FISHER, R.: The Use of Multiple Measurements in Taxonomic Problems. Annals of Eugenics, 7, 1, pp. 179–188, (1936).Google Scholar
  5. [Murphy, Aha, 97]
    MURPHY, Rand AHA, D. W.: UCI repository of machine learning databases-a machine-readable data repository. Maintained at the Department of Information and Computer Science, University of California, Irvine. Anonymous ftp from ftp.ics.uci.edu in the directory pub/machine-learning-databases, (1997).Google Scholar
  6. [Murthy, Kasif, Salzberg, 94]
    MURTHY, S. K., KASIF, S., & SALZBERG, S.: A system for induction of oblique decision trees. Journal of Artificial Intelligence Research, 2, pp. 1–32, (1994).Google Scholar
  7. [Quinlan, 93]
    QUINLAN, J. R.: C4.5: Programs for Machine Learning, p. x+302, Morgan Kaufmann Publishers, San Mateo (California), (1993).Google Scholar
  8. [Ranilla, Bahamonde, 95]
    RANILLA PASTOR, J. and BAHAMONDE, A., Segmentación de valores numéricos para el aprendizaje a partir de ejemplos. Proceedings of VI Conferencia de la Asociación Española para la Inteligencia Artificial, CAEPIA-95, pp. 225–234. Alicante (1995).Google Scholar
  9. [Ranilla, Mones, Bahamonde, 97]
    RANILLA, J.; MONES, R., and BAHAMONDE, A., El nivel de impureza de una regla de clasificación aprendida a partir de ejemplos. Proceedings of VII Conferencia de la Asociación Española para la Inteligencia Artificial, CAEPIA-97. Málaga, Spain, November, (1997).Google Scholar
  10. [Spiegel, 70]
    SPIEGEL, M. R.: Estadística, pp. x+357. McGraw-Hill, Atlacomulco, México, (1970).Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 1998

Authors and Affiliations

  • Oscar Luaces
    • 1
  • Jaime Alonso
    • 1
  • Enrique A. de la Cal
    • 1
  • José Ranilla
    • 1
  • Antonio Bahamonde
    • 1
  1. 1.Centro de Inteligencia ArtificialUniversidad de Oviedo en GijónGijónSpain

Personalised recommendations