Multi-objective optimization of base classifiers in StackingC by NSGA-II for intrusion detection

Michael Milliken, Yaxin Bi, Leo Galway, Glenn Hawe

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

3 Citations (Scopus)

Abstract

Multiple Classifier Systems are often found to be useful for improving individual results by combining a set of classifier decisions where a single base level classifier may not achieve the same level of results. However not every set of base classifiers improve results, therefore a selection of a set of classifiers is required. The process of selecting base level classifiers for a multiple classifier system may be performed by the use of a Genetic Algorithm. The aim of this work is the selection of optimal sets of base level classifies using an evolutionary computation approach. In addition, a comparative analysis is made of the performance of the generated ensembles against the individual base level classifiers.
LanguageEnglish
Title of host publicationUnknown Host Publication
Pages1-8
Number of pages8
DOIs
Publication statusE-pub ahead of print - 13 Feb 2017
Event2016 IEEE Symposium Series on Computational Intelligence (SSCI) - Athens, Greece
Duration: 13 Feb 2017 → …

Conference

Conference2016 IEEE Symposium Series on Computational Intelligence (SSCI)
Period13/02/17 → …

Fingerprint

Intrusion detection
Multiobjective optimization
Classifiers
Evolutionary algorithms
Genetic algorithms

Keywords

  • intrusion detection
  • cybersecurity
  • network security

Cite this

@inproceedings{053436528a8a4240805f15551a6329bf,
title = "Multi-objective optimization of base classifiers in StackingC by NSGA-II for intrusion detection",
abstract = "Multiple Classifier Systems are often found to be useful for improving individual results by combining a set of classifier decisions where a single base level classifier may not achieve the same level of results. However not every set of base classifiers improve results, therefore a selection of a set of classifiers is required. The process of selecting base level classifiers for a multiple classifier system may be performed by the use of a Genetic Algorithm. The aim of this work is the selection of optimal sets of base level classifies using an evolutionary computation approach. In addition, a comparative analysis is made of the performance of the generated ensembles against the individual base level classifiers.",
keywords = "intrusion detection, cybersecurity, network security",
author = "Michael Milliken and Yaxin Bi and Leo Galway and Glenn Hawe",
year = "2017",
month = "2",
day = "13",
doi = "10.1109/SSCI.2016.7849977",
language = "English",
isbn = "978-1-5090-4241-8",
pages = "1--8",
booktitle = "Unknown Host Publication",

}

Milliken, M, Bi, Y, Galway, L & Hawe, G 2017, Multi-objective optimization of base classifiers in StackingC by NSGA-II for intrusion detection. in Unknown Host Publication. pp. 1-8, 2016 IEEE Symposium Series on Computational Intelligence (SSCI), 13/02/17. https://doi.org/10.1109/SSCI.2016.7849977

Multi-objective optimization of base classifiers in StackingC by NSGA-II for intrusion detection. / Milliken, Michael; Bi, Yaxin; Galway, Leo; Hawe, Glenn.

Unknown Host Publication. 2017. p. 1-8.

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

TY - GEN

T1 - Multi-objective optimization of base classifiers in StackingC by NSGA-II for intrusion detection

AU - Milliken, Michael

AU - Bi, Yaxin

AU - Galway, Leo

AU - Hawe, Glenn

PY - 2017/2/13

Y1 - 2017/2/13

N2 - Multiple Classifier Systems are often found to be useful for improving individual results by combining a set of classifier decisions where a single base level classifier may not achieve the same level of results. However not every set of base classifiers improve results, therefore a selection of a set of classifiers is required. The process of selecting base level classifiers for a multiple classifier system may be performed by the use of a Genetic Algorithm. The aim of this work is the selection of optimal sets of base level classifies using an evolutionary computation approach. In addition, a comparative analysis is made of the performance of the generated ensembles against the individual base level classifiers.

AB - Multiple Classifier Systems are often found to be useful for improving individual results by combining a set of classifier decisions where a single base level classifier may not achieve the same level of results. However not every set of base classifiers improve results, therefore a selection of a set of classifiers is required. The process of selecting base level classifiers for a multiple classifier system may be performed by the use of a Genetic Algorithm. The aim of this work is the selection of optimal sets of base level classifies using an evolutionary computation approach. In addition, a comparative analysis is made of the performance of the generated ensembles against the individual base level classifiers.

KW - intrusion detection

KW - cybersecurity

KW - network security

U2 - 10.1109/SSCI.2016.7849977

DO - 10.1109/SSCI.2016.7849977

M3 - Conference contribution

SN - 978-1-5090-4241-8

SP - 1

EP - 8

BT - Unknown Host Publication

ER -