APPROXIMATE REASONING TO LEARN CLASSIFICATION RULES

Amel Borgi

2006

Abstract

In this paper, we propose an original use of approximate reasoning not only as a mode of inference but also as a means to refine a learning process. This work is done within the framework of the supervised learning method SUCRAGE which is based on automatic generation of classification rules. Production rules whose conclusions are accompanied by belief degrees, are obtained by supervised learning from a training set. These rules are then exploited by a basic inference engine: it fires only the rules with which the new observation to classify matches exactly. To introduce more flexibility, this engine was extended to an approximate inference which allows to fire rules not too far from the new observation. In this paper, we propose to use approximate reasoning to generate new rules with widened premises: thus imprecision of the observations are taken into account and problems due to the discretization of continuous attributes are eased. The objective is then to exploit the new base of rules by a basic inference engine, easier to interpret. The proposed method was implemented and experimental tests were carried out.

References

  1. Borgi A., 2005. Différentes méthodes pour optimiser le nombre de règles de classification dans SUCRAGE . 3rd Int. Conf. Sciences of Electronic, Technologies of Information and Telecom. SETIT 2005, 11 p., Tunisia.
  2. Borgi A, Akdag H., 2001. Apprentissage supervisé et raisonnement approximatif, l'hypothèse des imperfections. Revue d'Intelligence Artificielle, vol 15, n°1, pp 55-85, Editions Hermès.
  3. Borgi A., Akdag. H., 2001. Knowledge based supervised fuzzy-classification : An application to image processing. Annals of Mathematics and Artificial Intelligence, Vol 32, p 67-86.
  4. Borgi A., 1999. Apprentissage supervisé par génération de règles : le système SUCRAGE, Thèse de doctorat (PhD thesis), Université Paris VI.
  5. Duch W., R. Setiono, J.M. Zurada, 2004. Computational Intelligence Methods for Rule-Based Data Understanding. Proceedings of the IEEE, Vol. 92, 5.
  6. El-Sayed M., Pacholczyk D., 2003. Towards a Symbolic Interpretation of Approximate Reasoning. Electronic Notes in Theoretical Computer Science, Volume 82, Issue 4, Pages 1-12.
  7. Gupta M. M., Qi J., 1991. Connectives (And, Or, Not) and T-Operators in Fuzzy Reasoning. Conditional Inference and Logic for Intelligent Systems, 211-233.
  8. Haton J-P., Bouzid N., Charpillet F., Haton M., Lâasri B., Lâasri H., Marquis P., Mondot T., Napoli A., 1991 Le raisonnement en intelligence artificielle. InterEditions.
  9. Kohavi R., 1995. A Study of Cross-Validation and Bootstrap for Accuracy Estimation and Model Selection, Proc. of the Fourteenth International Joint Conference on Artificial Intelligence, Vol. 2.
  10. Michalski R.S., S. Ryszard, 1983. A theory and methodology of inductive learning. Machine Learning : An Artificial Intelligence Approach, Vol. I, 83-134.
  11. Nozaki K., Ishibuchi H., Tanaka H., 1994. Selecting Fuzzy Rules with Forgetting in Fuzzy Classification Systems. Proc. 3rd IEEE. Conf. Fuzzy Systems Vol. 1.
  12. Pearl. J., 1990. Numerical Uncertainty In Expert Systems. Readings in Uncertain Reasoning, Ed. by Shafer and Pearl. Morgan Kaufman publishers. California.
  13. Ruspini E., 1991. On the semantics of fuzzy logic. International Journal of Approximate Reasoning, 5.
  14. Vernazza G., 1993. Image Classification By Extended Certainty Factors. Pattern Recognition, vol. 26, n° 11, p. 1683-1694, Pergamon Press Ltd.
  15. Yager R.R., 2000. Approximate reasoning and conflict resolution. International Journal of Approximate Reasoning, 25, p. 15-42, Elsevier.
  16. Zadeh L.A., 1979. A Theory of Approximate Reasoning. Machine Intelligence, vol. 9, p. 149-194.
  17. Zhou Z. H., 2003. Three perspectives of data mining. Artificial Intelligence 143, 139-146, Elsevier.
Download


Paper Citation


in Harvard Style

Borgi A. (2006). APPROXIMATE REASONING TO LEARN CLASSIFICATION RULES . In Proceedings of the First International Conference on Software and Data Technologies - Volume 2: ICSOFT, ISBN 978-972-8865-69-6, pages 203-210. DOI: 10.5220/0001310402030210


in Bibtex Style

@conference{icsoft06,
author={Amel Borgi},
title={APPROXIMATE REASONING TO LEARN CLASSIFICATION RULES},
booktitle={Proceedings of the First International Conference on Software and Data Technologies - Volume 2: ICSOFT,},
year={2006},
pages={203-210},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0001310402030210},
isbn={978-972-8865-69-6},
}


in EndNote Style

TY - CONF
JO - Proceedings of the First International Conference on Software and Data Technologies - Volume 2: ICSOFT,
TI - APPROXIMATE REASONING TO LEARN CLASSIFICATION RULES
SN - 978-972-8865-69-6
AU - Borgi A.
PY - 2006
SP - 203
EP - 210
DO - 10.5220/0001310402030210