[48730] Artykuł: A multi-objective-genetic-optimization-based data-driven fuzzy classifier for technical applicationsCzasopismo: Proceedings of the 2016 IEEE 25th International Symposium on Industrial Electronics (ISIE) Strony: 78-83ISSN: 2163-5137 ISBN: 978-1-5090-0873-5 Wydawca: IEEE, 345 E 47TH ST, NEW YORK, NY 10017 USA Opublikowano: 2016 Seria wydawnicza: Proceedings of the IEEE International Symposium on Industrial Electronics Autorzy / Redaktorzy / Twórcy Grupa MNiSW: Materiały z konferencji międzynarodowej (zarejestrowane w Web of Science) Punkty MNiSW: 15 Klasyfikacja Web of Science: Proceedings Paper Pełny tekst DOI Web of Science |
The main contribution of this paper is our approach to automatic design from data a collection of fuzzy rule-based classifiers (FR-BCs) characterized by a high spread and well-balanced distribution of their accuracy-interpretability trade-offs. In order to achieve this goal, several multi-objective genetic optimization algorithms (M-OGOAs), including our original algorithm, have been applied. A general concept, a knowledge base, a fuzzy inference engine as well as main components of genetic learning and M-OGOA-based optimization of our FR-BCs are outlined. The proposed FR-BCs with genetically optimized accuracy-interpretability trade-off are effective and modern tools for intelligent decision support in various fields. The application to decision support in technical field of glass identification in forensic science is presented as an illustration of our approach.