Learning fuzzy classification rules from imbalanced datasets using multi-objective evolutionary algorithm

C. Edward Hinojosa; Heloisa A. Camargo; V. Yvan J. Tupac

doi:10.1109/LA-CCI.2015.7435959

Learning fuzzy classification rules from imbalanced datasets using multi-objective evolutionary algorithm

C. Edward Hinojosa, Heloisa A. Camargo, V. Yvan J. Tupac

Departamento Académico de Ingeniería de Sistemas e Informática

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

4 Scopus citations

Abstract

Fuzzy systems have been used to solve different types of problems, for example, classification problems. Genetic algorithms are a type of evolutionary algorithms used to automatically learn or tune components of the fuzzy systems from data. Recently multi-objective evolutionary algorithms have been used for this task, since they can consider multiple conflicting objectives, for example, accuracy and interpretability which are desirable properties of the fuzzy systems. Learning rules from imbalanced datasets is considered a research trend in this area. This work proposes a method to learn fuzzy classification rules from imbalanced datasets using multi-objective genetic algorithms and the iterative rule learning approach. In this approach, a single rule is learnt in each execution of the multi-objective evolutionary algorithm. The proposed method contains two phases: (i) pre-processing, to balance the imbalanced dataset; (ii) iterative fuzzy rule learning using multi-objective evolutionary algorithms. The algorithm uses two objectives: the accuracy and number of conditions of each fuzzy rule; the method proposed here is an extension of a method previously proposed by the authors. The results show that the new proposed method has a good performance. The obtained accuracy and the number of conditions are better than other genetic method used in the comparison analysis.

Original language	English
Title of host publication	2015 Latin-America Congress on Computational Intelligence, LA-CCI 2015
Editors	Marley M. B. R. Vellasco, Yvan J. Tupac Valdivia, Heitor Silverio Lopes
Publisher	Institute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)	9781467384186
DOIs	https://doi.org/10.1109/LA-CCI.2015.7435959
State	Published - 17 Mar 2016
Event	2nd Latin-America Congress on Computational Intelligence, LA-CCI 2015 - Curitiba, Brazil Duration: 13 Oct 2015 → 16 Oct 2015

Publication series

Name	2015 Latin-America Congress on Computational Intelligence, LA-CCI 2015

Conference

Conference	2nd Latin-America Congress on Computational Intelligence, LA-CCI 2015
Country/Territory	Brazil
City	Curitiba
Period	13/10/15 → 16/10/15

Bibliographical note

Publisher Copyright:
© 2015 IEEE.

Access to Document

10.1109/LA-CCI.2015.7435959

Cite this

Hinojosa, C. E., Camargo, H. A., & Tupac, V. Y. J. (2016). Learning fuzzy classification rules from imbalanced datasets using multi-objective evolutionary algorithm. In M. M. B. R. Vellasco, Y. J. Tupac Valdivia, & H. S. Lopes (Eds.), 2015 Latin-America Congress on Computational Intelligence, LA-CCI 2015 [7435959] (2015 Latin-America Congress on Computational Intelligence, LA-CCI 2015). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/LA-CCI.2015.7435959

Hinojosa, C. Edward ; Camargo, Heloisa A. ; Tupac, V. Yvan J. / Learning fuzzy classification rules from imbalanced datasets using multi-objective evolutionary algorithm. 2015 Latin-America Congress on Computational Intelligence, LA-CCI 2015. editor / Marley M. B. R. Vellasco ; Yvan J. Tupac Valdivia ; Heitor Silverio Lopes. Institute of Electrical and Electronics Engineers Inc., 2016. (2015 Latin-America Congress on Computational Intelligence, LA-CCI 2015).

@inproceedings{629f58964bf24400a68397a690af33cc,

title = "Learning fuzzy classification rules from imbalanced datasets using multi-objective evolutionary algorithm",

abstract = "Fuzzy systems have been used to solve different types of problems, for example, classification problems. Genetic algorithms are a type of evolutionary algorithms used to automatically learn or tune components of the fuzzy systems from data. Recently multi-objective evolutionary algorithms have been used for this task, since they can consider multiple conflicting objectives, for example, accuracy and interpretability which are desirable properties of the fuzzy systems. Learning rules from imbalanced datasets is considered a research trend in this area. This work proposes a method to learn fuzzy classification rules from imbalanced datasets using multi-objective genetic algorithms and the iterative rule learning approach. In this approach, a single rule is learnt in each execution of the multi-objective evolutionary algorithm. The proposed method contains two phases: (i) pre-processing, to balance the imbalanced dataset; (ii) iterative fuzzy rule learning using multi-objective evolutionary algorithms. The algorithm uses two objectives: the accuracy and number of conditions of each fuzzy rule; the method proposed here is an extension of a method previously proposed by the authors. The results show that the new proposed method has a good performance. The obtained accuracy and the number of conditions are better than other genetic method used in the comparison analysis.",

author = "Hinojosa, {C. Edward} and Camargo, {Heloisa A.} and Tupac, {V. Yvan J.}",

note = "Publisher Copyright: {\textcopyright} 2015 IEEE.; 2nd Latin-America Congress on Computational Intelligence, LA-CCI 2015 ; Conference date: 13-10-2015 Through 16-10-2015",

year = "2016",

month = mar,

day = "17",

doi = "10.1109/LA-CCI.2015.7435959",

language = "Ingl{\'e}s",

series = "2015 Latin-America Congress on Computational Intelligence, LA-CCI 2015",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

editor = "Vellasco, {Marley M. B. R.} and {Tupac Valdivia}, {Yvan J.} and Lopes, {Heitor Silverio}",

booktitle = "2015 Latin-America Congress on Computational Intelligence, LA-CCI 2015",

address = "Estados Unidos",

}

Hinojosa, CE, Camargo, HA & Tupac, VYJ 2016, Learning fuzzy classification rules from imbalanced datasets using multi-objective evolutionary algorithm. in MMBR Vellasco, YJ Tupac Valdivia & HS Lopes (eds), 2015 Latin-America Congress on Computational Intelligence, LA-CCI 2015., 7435959, 2015 Latin-America Congress on Computational Intelligence, LA-CCI 2015, Institute of Electrical and Electronics Engineers Inc., 2nd Latin-America Congress on Computational Intelligence, LA-CCI 2015, Curitiba, Brazil, 13/10/15. https://doi.org/10.1109/LA-CCI.2015.7435959

Learning fuzzy classification rules from imbalanced datasets using multi-objective evolutionary algorithm. / Hinojosa, C. Edward; Camargo, Heloisa A.; Tupac, V. Yvan J.
2015 Latin-America Congress on Computational Intelligence, LA-CCI 2015. ed. / Marley M. B. R. Vellasco; Yvan J. Tupac Valdivia; Heitor Silverio Lopes. Institute of Electrical and Electronics Engineers Inc., 2016. 7435959 (2015 Latin-America Congress on Computational Intelligence, LA-CCI 2015).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Learning fuzzy classification rules from imbalanced datasets using multi-objective evolutionary algorithm

AU - Hinojosa, C. Edward

AU - Camargo, Heloisa A.

AU - Tupac, V. Yvan J.

PY - 2016/3/17

Y1 - 2016/3/17

N2 - Fuzzy systems have been used to solve different types of problems, for example, classification problems. Genetic algorithms are a type of evolutionary algorithms used to automatically learn or tune components of the fuzzy systems from data. Recently multi-objective evolutionary algorithms have been used for this task, since they can consider multiple conflicting objectives, for example, accuracy and interpretability which are desirable properties of the fuzzy systems. Learning rules from imbalanced datasets is considered a research trend in this area. This work proposes a method to learn fuzzy classification rules from imbalanced datasets using multi-objective genetic algorithms and the iterative rule learning approach. In this approach, a single rule is learnt in each execution of the multi-objective evolutionary algorithm. The proposed method contains two phases: (i) pre-processing, to balance the imbalanced dataset; (ii) iterative fuzzy rule learning using multi-objective evolutionary algorithms. The algorithm uses two objectives: the accuracy and number of conditions of each fuzzy rule; the method proposed here is an extension of a method previously proposed by the authors. The results show that the new proposed method has a good performance. The obtained accuracy and the number of conditions are better than other genetic method used in the comparison analysis.

AB - Fuzzy systems have been used to solve different types of problems, for example, classification problems. Genetic algorithms are a type of evolutionary algorithms used to automatically learn or tune components of the fuzzy systems from data. Recently multi-objective evolutionary algorithms have been used for this task, since they can consider multiple conflicting objectives, for example, accuracy and interpretability which are desirable properties of the fuzzy systems. Learning rules from imbalanced datasets is considered a research trend in this area. This work proposes a method to learn fuzzy classification rules from imbalanced datasets using multi-objective genetic algorithms and the iterative rule learning approach. In this approach, a single rule is learnt in each execution of the multi-objective evolutionary algorithm. The proposed method contains two phases: (i) pre-processing, to balance the imbalanced dataset; (ii) iterative fuzzy rule learning using multi-objective evolutionary algorithms. The algorithm uses two objectives: the accuracy and number of conditions of each fuzzy rule; the method proposed here is an extension of a method previously proposed by the authors. The results show that the new proposed method has a good performance. The obtained accuracy and the number of conditions are better than other genetic method used in the comparison analysis.

UR - http://www.scopus.com/inward/record.url?scp=84969670087&partnerID=8YFLogxK

U2 - 10.1109/LA-CCI.2015.7435959

DO - 10.1109/LA-CCI.2015.7435959

M3 - Contribución a la conferencia

AN - SCOPUS:84969670087

T3 - 2015 Latin-America Congress on Computational Intelligence, LA-CCI 2015

BT - 2015 Latin-America Congress on Computational Intelligence, LA-CCI 2015

A2 - Vellasco, Marley M. B. R.

A2 - Tupac Valdivia, Yvan J.

A2 - Lopes, Heitor Silverio

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 2nd Latin-America Congress on Computational Intelligence, LA-CCI 2015

Y2 - 13 October 2015 through 16 October 2015

ER -

Hinojosa CE, Camargo HA, Tupac VYJ. Learning fuzzy classification rules from imbalanced datasets using multi-objective evolutionary algorithm. In Vellasco MMBR, Tupac Valdivia YJ, Lopes HS, editors, 2015 Latin-America Congress on Computational Intelligence, LA-CCI 2015. Institute of Electrical and Electronics Engineers Inc. 2016. 7435959. (2015 Latin-America Congress on Computational Intelligence, LA-CCI 2015). doi: 10.1109/LA-CCI.2015.7435959

Learning fuzzy classification rules from imbalanced datasets using multi-objective evolutionary algorithm

Abstract

Publication series

Conference

Bibliographical note

Access to Document

Other files and links

Fingerprint

Cite this