Adaptive resolution min-max classifier

Author

Rizzi, A. ; Mascioli, F. M Frattale ; Martinelli, G.

Author_Institution

INFOCOM Dept., Rome Univ., Italy

Volume

2

fYear

1998

fDate

4-9 May 1998

Firstpage

1435

Abstract

This paper presents a new neuro-fuzzy classifier, inspired by the Simpson´s (1992, 1993) min-max model. By relying on a constructive approach, it overcomes some undesired properties of the original min-max algorithm. In particular, training result does not depend on pattern presentation order and hyperbox expansion is not limited by a fixed maximum size, so that it is possible to have different covering resolutions. Consequently, the new algorithm yields less complex networks, thus increasing the generalization capability in accordance with learning theory paradigms. Several tests are presented for illustration

Keywords

fuzzy neural nets; generalisation (artificial intelligence); learning (artificial intelligence); minimax techniques; pattern classification; constructive algorithms; fuzzy neural networks; generalization; learning theory; min-max neural network; neural fuzzy classifier; pattern classification; Artificial intelligence; Classification algorithms; Clustering algorithms; Complex networks; Neural networks; Partitioning algorithms; Pattern recognition; Process control; Size control; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Fuzzy Systems Proceedings, 1998. IEEE World Congress on Computational Intelligence., The 1998 IEEE International Conference on

Conference_Location

Anchorage, AK

ISSN

1098-7584

Print_ISBN

0-7803-4863-X

Type

conf

DOI

10.1109/FUZZY.1998.686330

Filename

686330