Genetic Algorithm-Based Optimal Deep Neural Network for Detecting Network Intrusions

Sourav Adhikary; Md Musfique Anwar; Mohammad Jabed Morshed Chowdhury; Iqbal H. Sarker

doi:10.1007/978-981-19-2347-0_12

Genetic Algorithm-Based Optimal Deep Neural Network for Detecting Network Intrusions

Sourav Adhikary
, Md Musfique Anwar
, Mohammad Jabed Morshed Chowdhury
, Iqbal H. Sarker^*

^*Corresponding author for this work

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

1 Scopus citations

Abstract

Computer network attacks are evolving in parallel with the evolution of hardware and neural network architecture. Despite major advancements in network intrusion detection system (NIDS) technology, most implementations still depend on signature-based intrusion detection systems, which cannot identify unknown attacks. Deep learning can help NIDS to detect novel threats since it has a strong generalization ability. The deep neural network’s architecture has a significant impact on the model’s results. We propose a genetic algorithm-based model to find the optimal number of hidden layers and the number of neurons in each layer of the deep neural network (DNN) architecture for the network intrusion detection binary classification problem. Experimental results demonstrate that the proposed DNN architecture shows better performance than classical machine learning algorithms at a lower computational cost.

Original language	English
Title of host publication	Lecture Notes on Data Engineering and Communications Technologies
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	145-156
Number of pages	12
DOIs	https://doi.org/10.1007/978-981-19-2347-0_12
State	Published - 2022
Externally published	Yes

Publication series

Name	Lecture Notes on Data Engineering and Communications Technologies
Volume	132
ISSN (Print)	2367-4512
ISSN (Electronic)	2367-4520

Bibliographical note

Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

Keywords

Deep neural network
Genetic algorithm
Hidden layer
Intrusion detection
Optimal architecture

ASJC Scopus subject areas

Information Systems
Media Technology
Computer Science Applications
Computer Networks and Communications
Electrical and Electronic Engineering

Access to Document

10.1007/978-981-19-2347-0_12

Cite this

Adhikary, S., Anwar, M. M., Chowdhury, M. J. M., & Sarker, I. H. (2022). Genetic Algorithm-Based Optimal Deep Neural Network for Detecting Network Intrusions. In Lecture Notes on Data Engineering and Communications Technologies (pp. 145-156). (Lecture Notes on Data Engineering and Communications Technologies; Vol. 132). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-19-2347-0_12

Adhikary, Sourav ; Anwar, Md Musfique ; Chowdhury, Mohammad Jabed Morshed et al. / Genetic Algorithm-Based Optimal Deep Neural Network for Detecting Network Intrusions. Lecture Notes on Data Engineering and Communications Technologies. Springer Science and Business Media Deutschland GmbH, 2022. pp. 145-156 (Lecture Notes on Data Engineering and Communications Technologies).

@inbook{33706fe09c2b4bf98a151efd613dea29,

title = "Genetic Algorithm-Based Optimal Deep Neural Network for Detecting Network Intrusions",

abstract = "Computer network attacks are evolving in parallel with the evolution of hardware and neural network architecture. Despite major advancements in network intrusion detection system (NIDS) technology, most implementations still depend on signature-based intrusion detection systems, which cannot identify unknown attacks. Deep learning can help NIDS to detect novel threats since it has a strong generalization ability. The deep neural network{\textquoteright}s architecture has a significant impact on the model{\textquoteright}s results. We propose a genetic algorithm-based model to find the optimal number of hidden layers and the number of neurons in each layer of the deep neural network (DNN) architecture for the network intrusion detection binary classification problem. Experimental results demonstrate that the proposed DNN architecture shows better performance than classical machine learning algorithms at a lower computational cost.",

keywords = "Deep neural network, Genetic algorithm, Hidden layer, Intrusion detection, Optimal architecture",

author = "Sourav Adhikary and Anwar, \{Md Musfique\} and Chowdhury, \{Mohammad Jabed Morshed\} and Sarker, \{Iqbal H.\}",

note = "Publisher Copyright: {\textcopyright} 2022, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.",

year = "2022",

doi = "10.1007/978-981-19-2347-0\_12",

language = "English",

series = "Lecture Notes on Data Engineering and Communications Technologies",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "145--156",

booktitle = "Lecture Notes on Data Engineering and Communications Technologies",

address = "Germany",

}

Adhikary, S, Anwar, MM, Chowdhury, MJM & Sarker, IH 2022, Genetic Algorithm-Based Optimal Deep Neural Network for Detecting Network Intrusions. in Lecture Notes on Data Engineering and Communications Technologies. Lecture Notes on Data Engineering and Communications Technologies, vol. 132, Springer Science and Business Media Deutschland GmbH, pp. 145-156. https://doi.org/10.1007/978-981-19-2347-0_12

Genetic Algorithm-Based Optimal Deep Neural Network for Detecting Network Intrusions. / Adhikary, Sourav; Anwar, Md Musfique; Chowdhury, Mohammad Jabed Morshed et al.
Lecture Notes on Data Engineering and Communications Technologies. Springer Science and Business Media Deutschland GmbH, 2022. p. 145-156 (Lecture Notes on Data Engineering and Communications Technologies; Vol. 132).

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

TY - CHAP

T1 - Genetic Algorithm-Based Optimal Deep Neural Network for Detecting Network Intrusions

AU - Adhikary, Sourav

AU - Anwar, Md Musfique

AU - Chowdhury, Mohammad Jabed Morshed

AU - Sarker, Iqbal H.

PY - 2022

Y1 - 2022

N2 - Computer network attacks are evolving in parallel with the evolution of hardware and neural network architecture. Despite major advancements in network intrusion detection system (NIDS) technology, most implementations still depend on signature-based intrusion detection systems, which cannot identify unknown attacks. Deep learning can help NIDS to detect novel threats since it has a strong generalization ability. The deep neural network’s architecture has a significant impact on the model’s results. We propose a genetic algorithm-based model to find the optimal number of hidden layers and the number of neurons in each layer of the deep neural network (DNN) architecture for the network intrusion detection binary classification problem. Experimental results demonstrate that the proposed DNN architecture shows better performance than classical machine learning algorithms at a lower computational cost.

AB - Computer network attacks are evolving in parallel with the evolution of hardware and neural network architecture. Despite major advancements in network intrusion detection system (NIDS) technology, most implementations still depend on signature-based intrusion detection systems, which cannot identify unknown attacks. Deep learning can help NIDS to detect novel threats since it has a strong generalization ability. The deep neural network’s architecture has a significant impact on the model’s results. We propose a genetic algorithm-based model to find the optimal number of hidden layers and the number of neurons in each layer of the deep neural network (DNN) architecture for the network intrusion detection binary classification problem. Experimental results demonstrate that the proposed DNN architecture shows better performance than classical machine learning algorithms at a lower computational cost.

KW - Deep neural network

KW - Genetic algorithm

KW - Hidden layer

KW - Intrusion detection

KW - Optimal architecture

UR - https://www.scopus.com/pages/publications/85135504466

U2 - 10.1007/978-981-19-2347-0_12

DO - 10.1007/978-981-19-2347-0_12

M3 - Chapter

AN - SCOPUS:85135504466

T3 - Lecture Notes on Data Engineering and Communications Technologies

SP - 145

EP - 156

BT - Lecture Notes on Data Engineering and Communications Technologies

PB - Springer Science and Business Media Deutschland GmbH

ER -

Adhikary S, Anwar MM, Chowdhury MJM, Sarker IH. Genetic Algorithm-Based Optimal Deep Neural Network for Detecting Network Intrusions. In Lecture Notes on Data Engineering and Communications Technologies. Springer Science and Business Media Deutschland GmbH. 2022. p. 145-156. (Lecture Notes on Data Engineering and Communications Technologies). doi: 10.1007/978-981-19-2347-0_12