FUZZY CLASSIFIER TRAINING WITH ONLY MAIN COMPETITORS

Authors

  • S. D. Shtovba Vinnytsia National Technical University
  • A. V. Halushchak Vinnytsia National Technical University

Keywords:

classification, fuzzy knowledge base, training, training criteria, main competitors

Abstract

The tie "input—output" is described by linguistic «if—then» rules where antecedents contain fuzzy terms "low", "medium", "high" in the fuzzy classifiers. To enhance the correctness it is necessary to train fuzzy classifier on experimental data. There have been proposed new criteria for fuzzy classifier training that take into account the difference of fuzzy output only to the main competitors. When the classification is correct the main competitor of the decision is the class with the second largest degree of membership. In cases of misclassification erroneous decision is the main competitor to the correct class.

Computer experiments with the tuning up of a fuzzy classifier for UCI-problem of recognition of Italian wines showed a significant advantage of the new training criteria. New criteria of training can be used not only for tuning fuzzy classifiers but for some other models, such as neural networks.

Author Biographies

S. D. Shtovba, Vinnytsia National Technical University

Dr. Sc. (Eng.), Professor, Professor of the Chair of Computer Control Systems

A. V. Halushchak, Vinnytsia National Technical University

Assistant of the Chair of Computer Control Systems

References

1. Kuncheva L. I. Fuzzy classifier design : Studies in Fuzziness and Soft Computing / L. I. Kuncheva. — Berlin—Heidelberg : Springer-Verlag, 2000. — Vol. 49. — 314 p.
2. Штовба С. Д. Проектирование нечетких систем средствами MATLAB. — М. : Горячая линия – Телеком, 2007. — 288 с.
3. Ishibuchi H. Classification and modeling with linguistic information granules: advanced approaches advanced approaches to linguistic data mining / Ishibuchi H., Nakashima T., Nii M. — Berlin – Heidelberg : Springer-Verlag. 2005. — 307 p.
4. Rotshtein A. Design and tuning of fuzzy rule-based system for medical diagnosis. In «Fuzzy and Neuro-Fuzzy Systems in Medicine» / Rotshtein A. ; Eds.: Teodorescu N. H.. Kandel A. and Jain L. C.). Boca–Raton : CRC–Press, 1998. — P. 243—289.
5. Rudziński F. A multi-objective genetic optimization of interpretability-oriented fuzzy rule-based classifiers / Rudziński F. // Applied Soft Computing. — 2016. — Vol. 38. — P. 118—133.
6. Shtovba S. Tuning the fuzzy classification models with various learning criteria: the case of credit data classification / Shtovba S., Pankevich O., Dounias G. // Proc. of Inter. Conference on Fuzzy Sets and Soft Computing in Economics and Finance. St. Petersburg (Russia), 2004. — Vol. 1. — St. Petersburg : Russian Fuzzy Systems Association, 2004. — P. 103—110.
7. Штовба С. Д. Порівняння критеріїв навчання нечіткого класифікатора / С. Д. Штовба // Вісник Вінницького полі-технічного інституту. — 2007. — № 6. — С. 84—91.
8. Штовба С. Д. Анализ критериев обучения нечеткого классификатора / С. Д. Штовба, О. Д. Панкевич , А. В. Нагорна // Автоматика и вычислительная техника. — 2015. — № 3. — С. 5—16.
9. Ishibuchi H. Voting in fuzzy rule-based systems for pattern classification problems / Ishibuchi H., Nakashima T., Morisawa T. // Fuzzy Sets and Systems. — 1999. — Vol. 103, № 2. — P. 223—238.
10. Растригин Л. А. Адаптация сложных систем. Методы и приложения / Л. А. Растригин. — Рига : Зинатне, 1981. — 375 с.
11. Ishibuchi H. Fuzzy rule selection by multi-objective genetic local search algorithms and rule evaluation measures in data mining / Ishibuchi H., Yamamoto T. // Fuzzy Sets and Systems. — 2004. — Vol. 141, № 1. — P. 59—88.
12. Штовба С. Д. Обеспечение точности и прозрачности нечеткой модели Мамдани при обучении по эксперимен-тальным данным / С. Д. Штовба // Проблемы управления и информатики. — 2007. — № 4. — С. 102—114.

Downloads

Abstract views: 167

Published

2016-03-16

How to Cite

[1]
S. D. Shtovba and A. V. Halushchak, “FUZZY CLASSIFIER TRAINING WITH ONLY MAIN COMPETITORS”, Вісник ВПІ, no. 1, pp. 124–132, Mar. 2016.

Issue

No. 1 (2016)

Section

Information technologies and computer sciences

Metrics

Downloads

Download data is not yet available.

License

Authors who publish with this journal agree to the following terms:

  • Authors retain copyright and grant the journal right of first publication.
  • Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgment of its initial publication in this journal.
  • Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).