A method for fuzzy rules extraction directly from numerical data and its application to pattern classification

Author

Abe, Shigeto ; Lan, Ming-Shong

Author_Institution

Res. Lab., Hitachi Ltd., Ibaraki, Japan

Volume

3

Issue

1

fYear

1995

fDate

2/1/1995 12:00:00 AM

Firstpage

18

Lastpage

28

Abstract

In this paper, we discuss a new method for extracting fuzzy rules directly from numerical input-output data for pattern classification. Fuzzy rules with variable fuzzy regions are defined by activation hyperboxes which show the existence region of data for a class and inhibition hyperboxes which inhibit the existence of data for that class. These rules are extracted from numerical data by recursively resolving overlaps between two classes. Then, optimal input variables for the rules are determined using the number of extracted rules as a criterion. The method is compared with neural networks using the Fisher iris data and a license plate recognition system for various examples

Keywords

fuzzy systems; inference mechanisms; knowledge based systems; pattern classification; Fisher iris data; activation hyperboxes; fuzzy rules extraction; inhibition hyperboxes; license plate recognition system; numerical data; pattern classification; Backpropagation algorithms; Data mining; Fuzzy neural networks; Fuzzy systems; Input variables; Iris; Knowledge acquisition; Licenses; Neural networks; Pattern classification;

fLanguage

English

Journal_Title

Fuzzy Systems, IEEE Transactions on

Publisher

ieee

ISSN

1063-6706

Type

jour

DOI

10.1109/91.366565

Filename

366565