A Comparative Study: Effect of Feature Selection Methods on Detecting Botnet Attacks in IoT Devices by Using Deep Learning

Authors

  • Department of Computer Science, College of Computer Science and Mathematics, University of Thi-Qar, Iraq
  • Department of Computer Science, College of Computer Science and Mathematics, University of Thi-Qar, Iraq

DOI:

https://doi.org/10.32792/jeps.v13i2.297

Abstract

The Internet of Things, or IoT, is now an important technology that is the basis for various innovations
in intelligent environments, including smart homes and innovative healthcare. Due to their architecture,
many IoT devices suffer from security issues, leading to increased electronic threats targeting IoT
devices, facilitating abuse and lack of security control. The most common attack in IoT environments is
the botnet attack, which supports various criminal activities. In this study, This article aims to study
different methods for selecting features and comparing them to find the best possible strategies for
selecting and reducing features to detect botnet attacks in IoT devices. By using the UNSW-NB15
dataset, we will analyze the system's performance suggested to solve the classification issue. On the
UNSW-NB15 dataset, the results obtained using the LSTM, BRNN, and GRU classifiers were analyzed
to solve the binary classification issue and multiclass classification issues and compare the performance
of three different selection features (Correlation method, GNDO method, and Lasso method) of network
intrusion detection. The proposed system was evaluated by using different performance metrics and
comparing the various techniques to show better performance. The results showed with that a Filter
method (Correlation) for selecting features is better than other methods, and the model GRU in deep
learning got the highest accuracy, amounting to 92.71% and 78.62% for both binary and multiple
classifications, respectively. This study can potentially be applied in practical settings to detect real-time
network intrusions with a dynamic nature.

References

M. Wazzan and D. Algazzawi, O. Bamasaq, A. Albeshri, L.Cheng, . “Internet of Things

botnet detection approaches”in Analysis and recommendations for future research. Applied Sciences,

11, December 2021.

K .Ashton,. “That ‘internet of things’ thing. RFID journal , 97-114 22,July 2009.

J.Kim and M. Shim, S.Hong, Y. Shin, E.Choi, ” Intelligent detection of iot botnets using

machine learning and deep learning”. Applied Sciences, 7009 19 ,October, 2020 .

A.Cirne, and P.R. Sousa, J.S. Resende, L. Antunes, “IoT security certifications”in

Challenges and potential approaches”.in Computers & Security, 116, 102669 1,may 2022.

L.A.Tawalbeh, and F. Muheidat, M. Tawalbeh, , M. Quwaider, M. (2020). IoT Privacy and

security ” in Challenges and solutions. Applied Sciences, 10(12), 4102 12,October , 2020 .

N. Koroniotis, and `N. Moustafa, E. Sitnikova, J. Slay , “Towards developing network forensic

mechanism for botnet activities in the IoT based on machine learning techniques. In Mobile Networks

and Management, 9th International Conference, MONAMI 2017, Melbourne, Australia, December 13-

, 2017, Proceedings 9 (pp. 30-44). Springer International Publishing 2018 .

SonicWall cyber threat report. https://www.sonicwall.com/2019-cyber-threat-report/ (accessed 27

February 2023).

A. Muhammad, and M. Asad, A.R Javed, “Robust early stage botnet detection using machine

learning”. In 2020 International Conference on Cyber Warfare and Security (ICCWS) (pp. 1-6). IEEE,

October,2020.

M. .Lefoane and L. Ghafir, S. Kabir, I.U. Awan “Machine learning for botnet detection”in An

optimized feature selection approach. In The 5th International Conference on Future Networks &

Distributed Systems (pp. 195-200) , December 2020.

W.C Shi, and H.M . Sun, “ DeepBot: a time-based botnet detection with deep learning” in Soft

Computing, 24, 16605-16616, 2020 .

B. Nugraha, A. Nambiar and T. Bauschert, "Performance Evaluation of Botnet Detection using

Deep Learning Techniques," 2020 11th International Conference on Network of the Future (NoF),

Bordeaux, France, pp. 141-149, doi: 10.1109/NoF50125.2020.9249198 ,2020.

S. I. Popoola, R. Ande, B. Adebisi, G. Gui, M. Hammoudeh and O. Jogunola, "Federated Deep

Learning for Zero-Day Botnet Attack Detection in IoT-Edge Devices," in IEEE Internet of Things

Journal, vol. 9, no. 5, pp. 3930-3944, doi: 10.1109/JIOT.2021.3100755,1 March, 2022..

H. Alkahtani, and T.H. Aldhyani, “ Botnet attack detection by using CNN-LSTM model for

Internet of Things applications” in Security and Communication Networks, 23 January 2021..

I. Idrissi, and Boukabous, M. Azizi, O. Moussaoui, H. El Fadili “ Toward a deep

learning-based intrusion detection system for IoT against botnet attacks”. IAES International

Journal of Artificial Intelligence, 110,10, January 2021

T. Hasan et al., "Securing Industrial Internet of Things Against Botnet Attacks Using Hybrid

Deep Learning Approach," in IEEE Transactions on Network Science and Engineering, doi:

1109/TNSE.,.3168533,2022.

The UNSW-NB15 Dataset | UNSW Research, https://research.unsw.edu.au/projects/unsw-nb15-

dataset ,2015

I. Ahmad, and Q.E.Ul Haq ,M. Imran, M.O Alassafi,R.A. AlGhamdi.,“ An efficient

network intrusion detection and classification system. Mathematic”s, 530,10,marach,2022.

Z. Zoghi, and G. Serpen, . “Unsw-nb15 computer security dataset ” in Analysis through

visualization. arXiv preprint arXiv:2101.05067,2021

A.G. Karegowda and A.S. Manjunath, M.A. Jayaram,” Comparative study of attribute

selection using gain ratio and correlation-based feature selection” in International Journal of Information

Technology and Knowledge Management, 271-277, 2, February, 010.

.

P. Ghosh, and S. Azam, M. Jonkman, A.Karim,F.J.M. Shamrat, E. Ignatious, F. De Boer,”

Efficient prediction of cardiovascular disease using machine learning algorithms with relief and LASSO

feature selection techniques” in IEEE Access, 19304-19326, September 2021.

Downloads

Published

2023-07-13