Studying the robustness of ensembles of classifiers used for cancer diagnosis using microarrays datasets

Mohammed A. Gaafar, Noha Yousri, Mohamed A. Ismail

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

Abstract

Ensembles of classifiers were shown to provide better accuracy than single classifiers. However, the classification robustness is an important performance measure for classifiers and ensembles, besides accuracy, that should be considered. Increasing the robustness of classification systems results in reducing the probability of over-fitting. The robustness, as defined in this study, has not been studied in the literature. In this paper, a framework is used to prove that ensembles of classifiers are more robust than single classifiers. The framework selects different ensembles of classifiers and compares their robustness to the robustness of their members. The experiments performed on six different microarray datasets showed that ensembles of classifiers are more robust than their members.

Original languageEnglish
Title of host publicationProceedings - 2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013
Pages7-13
Number of pages7
DOIs
Publication statusPublished - 1 Dec 2013
Externally publishedYes
Event2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013 - Shanghai, China
Duration: 18 Dec 201321 Dec 2013

Other

Other2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013
CountryChina
CityShanghai
Period18/12/1321/12/13

Fingerprint

Microarrays
Classifiers
Experiments

Keywords

  • Cancer Classification
  • Classifiers Robustness
  • Ensemble Selection
  • Microarray Classification

ASJC Scopus subject areas

  • Biomedical Engineering

Cite this

Gaafar, M. A., Yousri, N., & Ismail, M. A. (2013). Studying the robustness of ensembles of classifiers used for cancer diagnosis using microarrays datasets. In Proceedings - 2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013 (pp. 7-13). [6732720] https://doi.org/10.1109/BIBM.2013.6732720

Studying the robustness of ensembles of classifiers used for cancer diagnosis using microarrays datasets. / Gaafar, Mohammed A.; Yousri, Noha; Ismail, Mohamed A.

Proceedings - 2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013. 2013. p. 7-13 6732720.

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

Gaafar, MA, Yousri, N & Ismail, MA 2013, Studying the robustness of ensembles of classifiers used for cancer diagnosis using microarrays datasets. in Proceedings - 2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013., 6732720, pp. 7-13, 2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013, Shanghai, China, 18/12/13. https://doi.org/10.1109/BIBM.2013.6732720
Gaafar MA, Yousri N, Ismail MA. Studying the robustness of ensembles of classifiers used for cancer diagnosis using microarrays datasets. In Proceedings - 2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013. 2013. p. 7-13. 6732720 https://doi.org/10.1109/BIBM.2013.6732720
Gaafar, Mohammed A. ; Yousri, Noha ; Ismail, Mohamed A. / Studying the robustness of ensembles of classifiers used for cancer diagnosis using microarrays datasets. Proceedings - 2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013. 2013. pp. 7-13
@inproceedings{ecb53cb544d74f8ea0862aa1383ffd0a,
title = "Studying the robustness of ensembles of classifiers used for cancer diagnosis using microarrays datasets",
abstract = "Ensembles of classifiers were shown to provide better accuracy than single classifiers. However, the classification robustness is an important performance measure for classifiers and ensembles, besides accuracy, that should be considered. Increasing the robustness of classification systems results in reducing the probability of over-fitting. The robustness, as defined in this study, has not been studied in the literature. In this paper, a framework is used to prove that ensembles of classifiers are more robust than single classifiers. The framework selects different ensembles of classifiers and compares their robustness to the robustness of their members. The experiments performed on six different microarray datasets showed that ensembles of classifiers are more robust than their members.",
keywords = "Cancer Classification, Classifiers Robustness, Ensemble Selection, Microarray Classification",
author = "Gaafar, {Mohammed A.} and Noha Yousri and Ismail, {Mohamed A.}",
year = "2013",
month = "12",
day = "1",
doi = "10.1109/BIBM.2013.6732720",
language = "English",
isbn = "9781479913091",
pages = "7--13",
booktitle = "Proceedings - 2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013",

}

TY - GEN

T1 - Studying the robustness of ensembles of classifiers used for cancer diagnosis using microarrays datasets

AU - Gaafar, Mohammed A.

AU - Yousri, Noha

AU - Ismail, Mohamed A.

PY - 2013/12/1

Y1 - 2013/12/1

N2 - Ensembles of classifiers were shown to provide better accuracy than single classifiers. However, the classification robustness is an important performance measure for classifiers and ensembles, besides accuracy, that should be considered. Increasing the robustness of classification systems results in reducing the probability of over-fitting. The robustness, as defined in this study, has not been studied in the literature. In this paper, a framework is used to prove that ensembles of classifiers are more robust than single classifiers. The framework selects different ensembles of classifiers and compares their robustness to the robustness of their members. The experiments performed on six different microarray datasets showed that ensembles of classifiers are more robust than their members.

AB - Ensembles of classifiers were shown to provide better accuracy than single classifiers. However, the classification robustness is an important performance measure for classifiers and ensembles, besides accuracy, that should be considered. Increasing the robustness of classification systems results in reducing the probability of over-fitting. The robustness, as defined in this study, has not been studied in the literature. In this paper, a framework is used to prove that ensembles of classifiers are more robust than single classifiers. The framework selects different ensembles of classifiers and compares their robustness to the robustness of their members. The experiments performed on six different microarray datasets showed that ensembles of classifiers are more robust than their members.

KW - Cancer Classification

KW - Classifiers Robustness

KW - Ensemble Selection

KW - Microarray Classification

UR - http://www.scopus.com/inward/record.url?scp=84894588395&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84894588395&partnerID=8YFLogxK

U2 - 10.1109/BIBM.2013.6732720

DO - 10.1109/BIBM.2013.6732720

M3 - Conference contribution

AN - SCOPUS:84894588395

SN - 9781479913091

SP - 7

EP - 13

BT - Proceedings - 2013 IEEE International Conference on Bioinformatics and Biomedicine, IEEE BIBM 2013

ER -