A Coevolutionary approach for classification problems: Preliminary results

Vu Van Truong, Bui Lam Thu, Thanh Nguyen

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

In the current classification problems, how to eliminate redundant features, find out important features, and choose appropriate classifiers for these feature set play a vital role. This paper presents a co-operative co-evolution approach (COCEA) with dual populations for optimizing both the artificial neural network (ANN) model and feature subset selection simultaneously. In COCEA, each feature subset is encoded as a binary string. Meanwhile, an ANN is represented in a matrix-form with real values of its weight and bias. During the process of evolution, a co-Operation mechanism is used to integrate the two populations and the final solution is the combination of the two most elite individuals of each population in a hope that the final solution will satisfy both ANN optimization and feature subset selection. The performance of COCEA is examined on both the well-known benchmark problems and Oil Spill dataset in SAR Images. In comparison with the other algorithms, experimental results illustrate that COCEA can significantly outperform other peer algorithms in terms of classification accuracy.

Original languageEnglish (US)
Title of host publicationNICS 2018 - Proceedings of 2018 5th NAFOSTED Conference on Information and Computer Science
EditorsHo Tu Bao, Le Anh Cuong, Ho Van Khuong, Bui Thu Lam
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages81-86
Number of pages6
ISBN (Electronic)9781538679838
DOIs
StatePublished - Jan 9 2019
Event5th NAFOSTED Conference on Information and Computer Science, NICS 2018 - Ho Chi Minh City, Viet Nam
Duration: Nov 23 2018Nov 24 2018

Publication series

NameNICS 2018 - Proceedings of 2018 5th NAFOSTED Conference on Information and Computer Science

Conference

Conference5th NAFOSTED Conference on Information and Computer Science, NICS 2018
CountryViet Nam
CityHo Chi Minh City
Period11/23/1811/24/18

Fingerprint

Neural networks
Oil spills
Set theory
Classifiers

Cite this

Van Truong, V., Lam Thu, B., & Nguyen, T. (2019). A Coevolutionary approach for classification problems: Preliminary results. In H. T. Bao, L. A. Cuong, H. V. Khuong, & B. T. Lam (Eds.), NICS 2018 - Proceedings of 2018 5th NAFOSTED Conference on Information and Computer Science (pp. 81-86). [8606876] (NICS 2018 - Proceedings of 2018 5th NAFOSTED Conference on Information and Computer Science). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/NICS.2018.8606876
Van Truong, Vu ; Lam Thu, Bui ; Nguyen, Thanh. / A Coevolutionary approach for classification problems : Preliminary results. NICS 2018 - Proceedings of 2018 5th NAFOSTED Conference on Information and Computer Science. editor / Ho Tu Bao ; Le Anh Cuong ; Ho Van Khuong ; Bui Thu Lam. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 81-86 (NICS 2018 - Proceedings of 2018 5th NAFOSTED Conference on Information and Computer Science).
@inproceedings{8e875323004f4c558bda3238b49e823e,
title = "A Coevolutionary approach for classification problems: Preliminary results",
abstract = "In the current classification problems, how to eliminate redundant features, find out important features, and choose appropriate classifiers for these feature set play a vital role. This paper presents a co-operative co-evolution approach (COCEA) with dual populations for optimizing both the artificial neural network (ANN) model and feature subset selection simultaneously. In COCEA, each feature subset is encoded as a binary string. Meanwhile, an ANN is represented in a matrix-form with real values of its weight and bias. During the process of evolution, a co-Operation mechanism is used to integrate the two populations and the final solution is the combination of the two most elite individuals of each population in a hope that the final solution will satisfy both ANN optimization and feature subset selection. The performance of COCEA is examined on both the well-known benchmark problems and Oil Spill dataset in SAR Images. In comparison with the other algorithms, experimental results illustrate that COCEA can significantly outperform other peer algorithms in terms of classification accuracy.",
author = "{Van Truong}, Vu and {Lam Thu}, Bui and Thanh Nguyen",
year = "2019",
month = "1",
day = "9",
doi = "https://doi.org/10.1109/NICS.2018.8606876",
language = "English (US)",
series = "NICS 2018 - Proceedings of 2018 5th NAFOSTED Conference on Information and Computer Science",
publisher = "Institute of Electrical and Electronics Engineers Inc.",
pages = "81--86",
editor = "Bao, {Ho Tu} and Cuong, {Le Anh} and Khuong, {Ho Van} and Lam, {Bui Thu}",
booktitle = "NICS 2018 - Proceedings of 2018 5th NAFOSTED Conference on Information and Computer Science",
address = "United States",

}

Van Truong, V, Lam Thu, B & Nguyen, T 2019, A Coevolutionary approach for classification problems: Preliminary results. in HT Bao, LA Cuong, HV Khuong & BT Lam (eds), NICS 2018 - Proceedings of 2018 5th NAFOSTED Conference on Information and Computer Science., 8606876, NICS 2018 - Proceedings of 2018 5th NAFOSTED Conference on Information and Computer Science, Institute of Electrical and Electronics Engineers Inc., pp. 81-86, 5th NAFOSTED Conference on Information and Computer Science, NICS 2018, Ho Chi Minh City, Viet Nam, 11/23/18. https://doi.org/10.1109/NICS.2018.8606876

A Coevolutionary approach for classification problems : Preliminary results. / Van Truong, Vu; Lam Thu, Bui; Nguyen, Thanh.

NICS 2018 - Proceedings of 2018 5th NAFOSTED Conference on Information and Computer Science. ed. / Ho Tu Bao; Le Anh Cuong; Ho Van Khuong; Bui Thu Lam. Institute of Electrical and Electronics Engineers Inc., 2019. p. 81-86 8606876 (NICS 2018 - Proceedings of 2018 5th NAFOSTED Conference on Information and Computer Science).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - A Coevolutionary approach for classification problems

T2 - Preliminary results

AU - Van Truong, Vu

AU - Lam Thu, Bui

AU - Nguyen, Thanh

PY - 2019/1/9

Y1 - 2019/1/9

N2 - In the current classification problems, how to eliminate redundant features, find out important features, and choose appropriate classifiers for these feature set play a vital role. This paper presents a co-operative co-evolution approach (COCEA) with dual populations for optimizing both the artificial neural network (ANN) model and feature subset selection simultaneously. In COCEA, each feature subset is encoded as a binary string. Meanwhile, an ANN is represented in a matrix-form with real values of its weight and bias. During the process of evolution, a co-Operation mechanism is used to integrate the two populations and the final solution is the combination of the two most elite individuals of each population in a hope that the final solution will satisfy both ANN optimization and feature subset selection. The performance of COCEA is examined on both the well-known benchmark problems and Oil Spill dataset in SAR Images. In comparison with the other algorithms, experimental results illustrate that COCEA can significantly outperform other peer algorithms in terms of classification accuracy.

AB - In the current classification problems, how to eliminate redundant features, find out important features, and choose appropriate classifiers for these feature set play a vital role. This paper presents a co-operative co-evolution approach (COCEA) with dual populations for optimizing both the artificial neural network (ANN) model and feature subset selection simultaneously. In COCEA, each feature subset is encoded as a binary string. Meanwhile, an ANN is represented in a matrix-form with real values of its weight and bias. During the process of evolution, a co-Operation mechanism is used to integrate the two populations and the final solution is the combination of the two most elite individuals of each population in a hope that the final solution will satisfy both ANN optimization and feature subset selection. The performance of COCEA is examined on both the well-known benchmark problems and Oil Spill dataset in SAR Images. In comparison with the other algorithms, experimental results illustrate that COCEA can significantly outperform other peer algorithms in terms of classification accuracy.

UR - http://www.scopus.com/inward/record.url?scp=85061923981&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85061923981&partnerID=8YFLogxK

U2 - https://doi.org/10.1109/NICS.2018.8606876

DO - https://doi.org/10.1109/NICS.2018.8606876

M3 - Conference contribution

T3 - NICS 2018 - Proceedings of 2018 5th NAFOSTED Conference on Information and Computer Science

SP - 81

EP - 86

BT - NICS 2018 - Proceedings of 2018 5th NAFOSTED Conference on Information and Computer Science

A2 - Bao, Ho Tu

A2 - Cuong, Le Anh

A2 - Khuong, Ho Van

A2 - Lam, Bui Thu

PB - Institute of Electrical and Electronics Engineers Inc.

ER -

Van Truong V, Lam Thu B, Nguyen T. A Coevolutionary approach for classification problems: Preliminary results. In Bao HT, Cuong LA, Khuong HV, Lam BT, editors, NICS 2018 - Proceedings of 2018 5th NAFOSTED Conference on Information and Computer Science. Institute of Electrical and Electronics Engineers Inc. 2019. p. 81-86. 8606876. (NICS 2018 - Proceedings of 2018 5th NAFOSTED Conference on Information and Computer Science). https://doi.org/10.1109/NICS.2018.8606876