Design of fuzzy rule-based classifier

Pruning and learning

Do Wan Kim, Jin Bae Park, Young Hoon Joo

Research output: Contribution to journalConference article

4 Citations (Scopus)

Abstract

This paper presents new pruning and learning methods for the fuzzy rule-based classifier. For the simplicity of the model structure, the unnecessary features for each fuzzy rule are eliminated through the iterative pruning algorithm. The quality of the feature is measured by the proposed correctness method, which is defined as the ratio of the fuzzy values for a set of the feature values on the decision region to one for all feature values. For the improvement of the classification performance, the parameters of the proposed classifier are adjusted by using the gradient descent method so that the misclassified feature vectors are correctly recategorized. Finally, the fuzzy rule-based classifier is tested on two data sets and is found to demonstrate an excellent performance.

Original languageEnglish
Pages (from-to)416-425
Number of pages10
JournalLecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science)
Volume3613
Issue numberPART I
Publication statusPublished - 2005 Oct 27

Fingerprint

Fuzzy rules
Fuzzy Rules
Pruning
Classifiers
Classifier
Gradient Descent Method
Model structures
Feature Vector
Simplicity
Correctness
Demonstrate
Learning
Design
Model

All Science Journal Classification (ASJC) codes

  • Hardware and Architecture

Cite this

@article{0126529c38144905ba77be64e55201dc,
title = "Design of fuzzy rule-based classifier: Pruning and learning",
abstract = "This paper presents new pruning and learning methods for the fuzzy rule-based classifier. For the simplicity of the model structure, the unnecessary features for each fuzzy rule are eliminated through the iterative pruning algorithm. The quality of the feature is measured by the proposed correctness method, which is defined as the ratio of the fuzzy values for a set of the feature values on the decision region to one for all feature values. For the improvement of the classification performance, the parameters of the proposed classifier are adjusted by using the gradient descent method so that the misclassified feature vectors are correctly recategorized. Finally, the fuzzy rule-based classifier is tested on two data sets and is found to demonstrate an excellent performance.",
author = "Kim, {Do Wan} and Park, {Jin Bae} and Joo, {Young Hoon}",
year = "2005",
month = "10",
day = "27",
language = "English",
volume = "3613",
pages = "416--425",
journal = "Lecture Notes in Computer Science",
issn = "0302-9743",
publisher = "Springer Verlag",
number = "PART I",

}

Design of fuzzy rule-based classifier : Pruning and learning. / Kim, Do Wan; Park, Jin Bae; Joo, Young Hoon.

In: Lecture Notes in Artificial Intelligence (Subseries of Lecture Notes in Computer Science), Vol. 3613, No. PART I, 27.10.2005, p. 416-425.

Research output: Contribution to journalConference article

TY - JOUR

T1 - Design of fuzzy rule-based classifier

T2 - Pruning and learning

AU - Kim, Do Wan

AU - Park, Jin Bae

AU - Joo, Young Hoon

PY - 2005/10/27

Y1 - 2005/10/27

N2 - This paper presents new pruning and learning methods for the fuzzy rule-based classifier. For the simplicity of the model structure, the unnecessary features for each fuzzy rule are eliminated through the iterative pruning algorithm. The quality of the feature is measured by the proposed correctness method, which is defined as the ratio of the fuzzy values for a set of the feature values on the decision region to one for all feature values. For the improvement of the classification performance, the parameters of the proposed classifier are adjusted by using the gradient descent method so that the misclassified feature vectors are correctly recategorized. Finally, the fuzzy rule-based classifier is tested on two data sets and is found to demonstrate an excellent performance.

AB - This paper presents new pruning and learning methods for the fuzzy rule-based classifier. For the simplicity of the model structure, the unnecessary features for each fuzzy rule are eliminated through the iterative pruning algorithm. The quality of the feature is measured by the proposed correctness method, which is defined as the ratio of the fuzzy values for a set of the feature values on the decision region to one for all feature values. For the improvement of the classification performance, the parameters of the proposed classifier are adjusted by using the gradient descent method so that the misclassified feature vectors are correctly recategorized. Finally, the fuzzy rule-based classifier is tested on two data sets and is found to demonstrate an excellent performance.

UR - http://www.scopus.com/inward/record.url?scp=26944485495&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=26944485495&partnerID=8YFLogxK

M3 - Conference article

VL - 3613

SP - 416

EP - 425

JO - Lecture Notes in Computer Science

JF - Lecture Notes in Computer Science

SN - 0302-9743

IS - PART I

ER -