Ensemble Learning utilising Feature Pairings for Intrusion Detection

Michael Milliken, Yaxin Bi, Leo Galway, Glenn Hawe

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

1 Citation (Scopus)

Abstract

Network intrusions may illicitly retrieve data/information, or prevent legitimate access. Reliable detection of network intrusions is an important problem, misclassification of an intrusion is an issue in and of itself reducing overall accuracy of detection. A variety of potential methods exist to develop an improved system to perform classification more accurately. Feature selection is one potential area that may be utilized to successfully improve performance by initially identifying sets and subsets of features that are relevant and nonredundant. Within this paper explicit pairings of features have been investigated in order to determine if the presence of pairings has a positive effect on classification, potentially increasing the accuracy of detecting intrusions correctly. In particular, classification using the ensemble algorithm, StackingC, with F-Measure performance and derived Information Gain Ratio, as well as their subsequent correlation as a combined measure, is presented.
LanguageEnglish
Title of host publicationUnknown Host Publication
Pages24-31
Number of pages7
Publication statusPublished - Oct 2015
EventWorld Congress on Internet Security (WorldCIS-2015) - Dublin, Ireland
Duration: 1 Oct 2015 → …

Conference

ConferenceWorld Congress on Internet Security (WorldCIS-2015)
Period1/10/15 → …

Fingerprint

Intrusion detection
Feature extraction

Keywords

  • Intrusion Detection
  • Ensemble Learning
  • Feature Selection

Cite this

@inproceedings{6a3ba549f8494133a3b7c1eb645897aa,
title = "Ensemble Learning utilising Feature Pairings for Intrusion Detection",
abstract = "Network intrusions may illicitly retrieve data/information, or prevent legitimate access. Reliable detection of network intrusions is an important problem, misclassification of an intrusion is an issue in and of itself reducing overall accuracy of detection. A variety of potential methods exist to develop an improved system to perform classification more accurately. Feature selection is one potential area that may be utilized to successfully improve performance by initially identifying sets and subsets of features that are relevant and nonredundant. Within this paper explicit pairings of features have been investigated in order to determine if the presence of pairings has a positive effect on classification, potentially increasing the accuracy of detecting intrusions correctly. In particular, classification using the ensemble algorithm, StackingC, with F-Measure performance and derived Information Gain Ratio, as well as their subsequent correlation as a combined measure, is presented.",
keywords = "Intrusion Detection, Ensemble Learning, Feature Selection",
author = "Michael Milliken and Yaxin Bi and Leo Galway and Glenn Hawe",
year = "2015",
month = "10",
language = "English",
pages = "24--31",
booktitle = "Unknown Host Publication",

}

Milliken, M, Bi, Y, Galway, L & Hawe, G 2015, Ensemble Learning utilising Feature Pairings for Intrusion Detection. in Unknown Host Publication. pp. 24-31, World Congress on Internet Security (WorldCIS-2015), 1/10/15.

Ensemble Learning utilising Feature Pairings for Intrusion Detection. / Milliken, Michael; Bi, Yaxin; Galway, Leo; Hawe, Glenn.

Unknown Host Publication. 2015. p. 24-31.

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

TY - GEN

T1 - Ensemble Learning utilising Feature Pairings for Intrusion Detection

AU - Milliken, Michael

AU - Bi, Yaxin

AU - Galway, Leo

AU - Hawe, Glenn

PY - 2015/10

Y1 - 2015/10

N2 - Network intrusions may illicitly retrieve data/information, or prevent legitimate access. Reliable detection of network intrusions is an important problem, misclassification of an intrusion is an issue in and of itself reducing overall accuracy of detection. A variety of potential methods exist to develop an improved system to perform classification more accurately. Feature selection is one potential area that may be utilized to successfully improve performance by initially identifying sets and subsets of features that are relevant and nonredundant. Within this paper explicit pairings of features have been investigated in order to determine if the presence of pairings has a positive effect on classification, potentially increasing the accuracy of detecting intrusions correctly. In particular, classification using the ensemble algorithm, StackingC, with F-Measure performance and derived Information Gain Ratio, as well as their subsequent correlation as a combined measure, is presented.

AB - Network intrusions may illicitly retrieve data/information, or prevent legitimate access. Reliable detection of network intrusions is an important problem, misclassification of an intrusion is an issue in and of itself reducing overall accuracy of detection. A variety of potential methods exist to develop an improved system to perform classification more accurately. Feature selection is one potential area that may be utilized to successfully improve performance by initially identifying sets and subsets of features that are relevant and nonredundant. Within this paper explicit pairings of features have been investigated in order to determine if the presence of pairings has a positive effect on classification, potentially increasing the accuracy of detecting intrusions correctly. In particular, classification using the ensemble algorithm, StackingC, with F-Measure performance and derived Information Gain Ratio, as well as their subsequent correlation as a combined measure, is presented.

KW - Intrusion Detection

KW - Ensemble Learning

KW - Feature Selection

M3 - Conference contribution

SP - 24

EP - 31

BT - Unknown Host Publication

ER -