An improved intrusion detection approach using synthetic minority over-sampling technique and deep belief network

S. Hasan Adil; S. Saad Azhar Ali; Kamran Raza; A. Mahmood Hussaan

doi:10.3233/978-1-61499-434-3-94

An improved intrusion detection approach using synthetic minority over-sampling technique and deep belief network

S. Hasan Adil^*
, S. Saad Azhar Ali
, Kamran Raza
, A. Mahmood Hussaan

^*Corresponding author for this work

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

9 Scopus citations

Abstract

This paper presents a network intrusion detection technique based on Synthetic Minority Over-Sampling Technique (SMOTE) and Deep Belief Network (DBN) applied to a class imbalance KDD-99 dataset. SMOTE is used to eliminate the class imbalance problem while intrusion classification is performed using DBN. The proposed technique first resolves the class imbalance problem in the KDD-99 dataset followed by DBN to estimate the initial model. The accuracy is further enhanced by using multilayer perceptron networks. The obtained results are compared with the existing best technique based on reduced size recurrent neural network. The study shows that our approach is competitive and efficient in classifying both intrusion and normal patterns in KDD-99 dataset.

Original language	English
Title of host publication	New Trends in Software Methodologies, Tools and Techniques - Proceedings of the 13th SoMeT 2014
Editors	Hamido Fujita, Ali Selamat, Habibollah Haron
Publisher	IOS Press BV
Pages	94-102
Number of pages	9
ISBN (Electronic)	9781614994336
DOIs	https://doi.org/10.3233/978-1-61499-434-3-94
State	Published - 2014
Externally published	Yes
Event	13th International Conference on New Trends in Intelligent Software Methodology Tools, and Techniques, SoMeT 2014 - Langkawi, Malaysia Duration: 22 Sep 2014 → 24 Sep 2014

Publication series

Name	Frontiers in Artificial Intelligence and Applications
Volume	265
ISSN (Print)	0922-6389
ISSN (Electronic)	1879-8314

Conference

Conference	13th International Conference on New Trends in Intelligent Software Methodology Tools, and Techniques, SoMeT 2014
Country/Territory	Malaysia
City	Langkawi
Period	22/09/14 → 24/09/14

Bibliographical note

Publisher Copyright:
© 2014 The authors and IOS Press. All rights reserved.

Keywords

DBN
Intrusion detection
Multi-layer perceptron

ASJC Scopus subject areas

Artificial Intelligence

Access to Document

10.3233/978-1-61499-434-3-94

Cite this

Adil, S. H., Ali, S. S. A., Raza, K., & Hussaan, A. M. (2014). An improved intrusion detection approach using synthetic minority over-sampling technique and deep belief network. In H. Fujita, A. Selamat, & H. Haron (Eds.), New Trends in Software Methodologies, Tools and Techniques - Proceedings of the 13th SoMeT 2014 (pp. 94-102). (Frontiers in Artificial Intelligence and Applications; Vol. 265). IOS Press BV. https://doi.org/10.3233/978-1-61499-434-3-94

Adil, S. Hasan ; Ali, S. Saad Azhar ; Raza, Kamran et al. / An improved intrusion detection approach using synthetic minority over-sampling technique and deep belief network. New Trends in Software Methodologies, Tools and Techniques - Proceedings of the 13th SoMeT 2014. editor / Hamido Fujita ; Ali Selamat ; Habibollah Haron. IOS Press BV, 2014. pp. 94-102 (Frontiers in Artificial Intelligence and Applications).

@inproceedings{8b564c377304422987c5951b579b6a17,

title = "An improved intrusion detection approach using synthetic minority over-sampling technique and deep belief network",

abstract = "This paper presents a network intrusion detection technique based on Synthetic Minority Over-Sampling Technique (SMOTE) and Deep Belief Network (DBN) applied to a class imbalance KDD-99 dataset. SMOTE is used to eliminate the class imbalance problem while intrusion classification is performed using DBN. The proposed technique first resolves the class imbalance problem in the KDD-99 dataset followed by DBN to estimate the initial model. The accuracy is further enhanced by using multilayer perceptron networks. The obtained results are compared with the existing best technique based on reduced size recurrent neural network. The study shows that our approach is competitive and efficient in classifying both intrusion and normal patterns in KDD-99 dataset.",

keywords = "DBN, Intrusion detection, Multi-layer perceptron",

author = "Adil, \{S. Hasan\} and Ali, \{S. Saad Azhar\} and Kamran Raza and Hussaan, \{A. Mahmood\}",

note = "Publisher Copyright: {\textcopyright} 2014 The authors and IOS Press. All rights reserved.; 13th International Conference on New Trends in Intelligent Software Methodology Tools, and Techniques, SoMeT 2014 ; Conference date: 22-09-2014 Through 24-09-2014",

year = "2014",

doi = "10.3233/978-1-61499-434-3-94",

language = "English",

series = "Frontiers in Artificial Intelligence and Applications",

publisher = "IOS Press BV",

pages = "94--102",

editor = "Hamido Fujita and Ali Selamat and Habibollah Haron",

booktitle = "New Trends in Software Methodologies, Tools and Techniques - Proceedings of the 13th SoMeT 2014",

address = "Netherlands",

}

Adil, SH, Ali, SSA, Raza, K & Hussaan, AM 2014, An improved intrusion detection approach using synthetic minority over-sampling technique and deep belief network. in H Fujita, A Selamat & H Haron (eds), New Trends in Software Methodologies, Tools and Techniques - Proceedings of the 13th SoMeT 2014. Frontiers in Artificial Intelligence and Applications, vol. 265, IOS Press BV, pp. 94-102, 13th International Conference on New Trends in Intelligent Software Methodology Tools, and Techniques, SoMeT 2014, Langkawi, Malaysia, 22/09/14. https://doi.org/10.3233/978-1-61499-434-3-94

An improved intrusion detection approach using synthetic minority over-sampling technique and deep belief network. / Adil, S. Hasan; Ali, S. Saad Azhar; Raza, Kamran et al.
New Trends in Software Methodologies, Tools and Techniques - Proceedings of the 13th SoMeT 2014. ed. / Hamido Fujita; Ali Selamat; Habibollah Haron. IOS Press BV, 2014. p. 94-102 (Frontiers in Artificial Intelligence and Applications; Vol. 265).

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

TY - GEN

T1 - An improved intrusion detection approach using synthetic minority over-sampling technique and deep belief network

AU - Adil, S. Hasan

AU - Ali, S. Saad Azhar

AU - Raza, Kamran

AU - Hussaan, A. Mahmood

PY - 2014

Y1 - 2014

N2 - This paper presents a network intrusion detection technique based on Synthetic Minority Over-Sampling Technique (SMOTE) and Deep Belief Network (DBN) applied to a class imbalance KDD-99 dataset. SMOTE is used to eliminate the class imbalance problem while intrusion classification is performed using DBN. The proposed technique first resolves the class imbalance problem in the KDD-99 dataset followed by DBN to estimate the initial model. The accuracy is further enhanced by using multilayer perceptron networks. The obtained results are compared with the existing best technique based on reduced size recurrent neural network. The study shows that our approach is competitive and efficient in classifying both intrusion and normal patterns in KDD-99 dataset.

AB - This paper presents a network intrusion detection technique based on Synthetic Minority Over-Sampling Technique (SMOTE) and Deep Belief Network (DBN) applied to a class imbalance KDD-99 dataset. SMOTE is used to eliminate the class imbalance problem while intrusion classification is performed using DBN. The proposed technique first resolves the class imbalance problem in the KDD-99 dataset followed by DBN to estimate the initial model. The accuracy is further enhanced by using multilayer perceptron networks. The obtained results are compared with the existing best technique based on reduced size recurrent neural network. The study shows that our approach is competitive and efficient in classifying both intrusion and normal patterns in KDD-99 dataset.

KW - DBN

KW - Intrusion detection

KW - Multi-layer perceptron

UR - https://www.scopus.com/pages/publications/84948783277

U2 - 10.3233/978-1-61499-434-3-94

DO - 10.3233/978-1-61499-434-3-94

M3 - Conference contribution

AN - SCOPUS:84948783277

T3 - Frontiers in Artificial Intelligence and Applications

SP - 94

EP - 102

BT - New Trends in Software Methodologies, Tools and Techniques - Proceedings of the 13th SoMeT 2014

A2 - Fujita, Hamido

A2 - Selamat, Ali

A2 - Haron, Habibollah

PB - IOS Press BV

T2 - 13th International Conference on New Trends in Intelligent Software Methodology Tools, and Techniques, SoMeT 2014

Y2 - 22 September 2014 through 24 September 2014

ER -

Adil SH, Ali SSA, Raza K, Hussaan AM. An improved intrusion detection approach using synthetic minority over-sampling technique and deep belief network. In Fujita H, Selamat A, Haron H, editors, New Trends in Software Methodologies, Tools and Techniques - Proceedings of the 13th SoMeT 2014. IOS Press BV. 2014. p. 94-102. (Frontiers in Artificial Intelligence and Applications). doi: 10.3233/978-1-61499-434-3-94

An improved intrusion detection approach using synthetic minority over-sampling technique and deep belief network

Abstract

Publication series

Conference

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Fingerprint

Cite this