Induction of fuzzy-rule-based classifiers with evolutionary boosting algorithms

This paper proposes a novel Adaboost algorithm to learn fuzzy-rule-based classifiers. Connections between iterative learning and boosting are analyzed in terms of their respective structures and the manner these algorithms address the cooperation-competition problem. The results are used to explain...

Full description

Saved in:
Bibliographic Details
Published in:IEEE transactions on fuzzy systems 2004-06, Vol.12 (3), p.296-308
Main Authors: del Jesus, M.J., Hoffmann, F., Navascues, L.J., Sanchez, L.
Format: Article
Language:English
Subjects:
Algorithm design and analysis
Algorithms
Approximation
Boosting
Classification
Classifiers
Computer science
Evolutionary
Evolutionary computation
Fuzzy
Fuzzy logic
Fuzzy sets
Genetics
Iterative algorithms
Tasks
Voting
Citations: Items that this one cites
Items that cite this one
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:This paper proposes a novel Adaboost algorithm to learn fuzzy-rule-based classifiers. Connections between iterative learning and boosting are analyzed in terms of their respective structures and the manner these algorithms address the cooperation-competition problem. The results are used to explain some properties of the former method. The evolutionary boosting scheme is applied to approximate and descriptive fuzzy-rule bases. The advantages of boosting fuzzy rules are assessed by performance comparisons between the proposed method and other classification schemes applied on a set of benchmark classification tasks.
ISSN:1063-6706
1941-0034
DOI:10.1109/TFUZZ.2004.825972