Re-Evaluating Deep Learning Attacks and Defenses in Cybersecurity Systems

Cybersecurity attacks pose a significant threat to the security of network systems through intrusions and illegal communications. Measuring the vulnerability of cybersecurity is crucial for refining the overall system security to further mitigate potential security risks. Machine learning (ML)-based...

Full description

Saved in:
Bibliographic Details
Main Authors: Meaad Ahmed, Qutaiba Alasad, Jiann-Shiun Yuan, Mohammed Alawad
Format: Article
Language:English
Published: MDPI AG 2024-12-01
Series:Big Data and Cognitive Computing
Subjects:
Online Access:https://www.mdpi.com/2504-2289/8/12/191
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1846105819632893952
author Meaad Ahmed
Qutaiba Alasad
Jiann-Shiun Yuan
Mohammed Alawad
author_facet Meaad Ahmed
Qutaiba Alasad
Jiann-Shiun Yuan
Mohammed Alawad
author_sort Meaad Ahmed
collection DOAJ
description Cybersecurity attacks pose a significant threat to the security of network systems through intrusions and illegal communications. Measuring the vulnerability of cybersecurity is crucial for refining the overall system security to further mitigate potential security risks. Machine learning (ML)-based intrusion detection systems (IDSs) are mainly designed to detect malicious network traffic. Unfortunately, ML models have recently been demonstrated to be vulnerable to adversarial perturbation, and therefore enable potential attackers to crash the system during normal operation. Among different attacks, generative adversarial networks (GANs) have been known as one of the most powerful threats to cybersecurity systems. To address these concerns, it is important to explore new defense methods and understand the nature of different types of attacks. In this paper, we investigate four serious attacks, GAN, Zeroth-Order Optimization (ZOO), kernel density estimation (KDE), and DeepFool attacks, on cybersecurity. Deep analysis was conducted on these attacks using three different cybersecurity datasets, ADFA-LD, CSE-CICIDS2018, and CSE-CICIDS2019. Our results have shown that KDE and DeepFool attacks are stronger than GANs in terms of attack success rate and impact on system performance. To demonstrate the effectiveness of our approach, we develop a defensive model using adversarial training where the DeepFool method is used to generate adversarial examples. The model is evaluated against GAN, ZOO, KDE, and DeepFool attacks to assess the level of system protection against adversarial perturbations. The experiment was conducted by leveraging a deep learning model as a classifier with the three aforementioned datasets. The results indicate that the proposed defensive model refines the resilience of the system and mitigates the presented serious attacks.
format Article
id doaj-art-f1e9059fee1949d88c4c41dd7db0d7b6
institution Kabale University
issn 2504-2289
language English
publishDate 2024-12-01
publisher MDPI AG
record_format Article
series Big Data and Cognitive Computing
spelling doaj-art-f1e9059fee1949d88c4c41dd7db0d7b62024-12-27T14:10:49ZengMDPI AGBig Data and Cognitive Computing2504-22892024-12-0181219110.3390/bdcc8120191Re-Evaluating Deep Learning Attacks and Defenses in Cybersecurity SystemsMeaad Ahmed0Qutaiba Alasad1Jiann-Shiun Yuan2Mohammed Alawad3Department of Computer Science, Tikrit University, Al-Qadesiyya District, Tikrit 34001, IraqDepartment of Cybersecurity, Tikrit University, Al-Qadesiyya District, Tikrit 34001, IraqDepartment of Electrical and Computer Engineering, University of Central Florida, Orlando, FL 32816, USADepartment of Electrical and Computer Engineering, Wayne State University, Detroit, MI 48202, USACybersecurity attacks pose a significant threat to the security of network systems through intrusions and illegal communications. Measuring the vulnerability of cybersecurity is crucial for refining the overall system security to further mitigate potential security risks. Machine learning (ML)-based intrusion detection systems (IDSs) are mainly designed to detect malicious network traffic. Unfortunately, ML models have recently been demonstrated to be vulnerable to adversarial perturbation, and therefore enable potential attackers to crash the system during normal operation. Among different attacks, generative adversarial networks (GANs) have been known as one of the most powerful threats to cybersecurity systems. To address these concerns, it is important to explore new defense methods and understand the nature of different types of attacks. In this paper, we investigate four serious attacks, GAN, Zeroth-Order Optimization (ZOO), kernel density estimation (KDE), and DeepFool attacks, on cybersecurity. Deep analysis was conducted on these attacks using three different cybersecurity datasets, ADFA-LD, CSE-CICIDS2018, and CSE-CICIDS2019. Our results have shown that KDE and DeepFool attacks are stronger than GANs in terms of attack success rate and impact on system performance. To demonstrate the effectiveness of our approach, we develop a defensive model using adversarial training where the DeepFool method is used to generate adversarial examples. The model is evaluated against GAN, ZOO, KDE, and DeepFool attacks to assess the level of system protection against adversarial perturbations. The experiment was conducted by leveraging a deep learning model as a classifier with the three aforementioned datasets. The results indicate that the proposed defensive model refines the resilience of the system and mitigates the presented serious attacks.https://www.mdpi.com/2504-2289/8/12/191network-based intrusion detection systemszeroth-order optimizationkernel density estimationgenerative adversarial networksDeepFooldeep learning
spellingShingle Meaad Ahmed
Qutaiba Alasad
Jiann-Shiun Yuan
Mohammed Alawad
Re-Evaluating Deep Learning Attacks and Defenses in Cybersecurity Systems
Big Data and Cognitive Computing
network-based intrusion detection systems
zeroth-order optimization
kernel density estimation
generative adversarial networks
DeepFool
deep learning
title Re-Evaluating Deep Learning Attacks and Defenses in Cybersecurity Systems
title_full Re-Evaluating Deep Learning Attacks and Defenses in Cybersecurity Systems
title_fullStr Re-Evaluating Deep Learning Attacks and Defenses in Cybersecurity Systems
title_full_unstemmed Re-Evaluating Deep Learning Attacks and Defenses in Cybersecurity Systems
title_short Re-Evaluating Deep Learning Attacks and Defenses in Cybersecurity Systems
title_sort re evaluating deep learning attacks and defenses in cybersecurity systems
topic network-based intrusion detection systems
zeroth-order optimization
kernel density estimation
generative adversarial networks
DeepFool
deep learning
url https://www.mdpi.com/2504-2289/8/12/191
work_keys_str_mv AT meaadahmed reevaluatingdeeplearningattacksanddefensesincybersecuritysystems
AT qutaibaalasad reevaluatingdeeplearningattacksanddefensesincybersecuritysystems
AT jiannshiunyuan reevaluatingdeeplearningattacksanddefensesincybersecuritysystems
AT mohammedalawad reevaluatingdeeplearningattacksanddefensesincybersecuritysystems