Investigation of performance of machine learning methods for cyber-attack detection

Abstract

The increase in internet based devices gets security problems in cyber environment. Malwares can disturb the functioning of systems and compromise data privacy due to vulnerabilities in systems. Intrusion Detection Systems (STS) are improved to determine and classify attacks. Artificial intelligence-based methods are used more frequently to improve STS systems. In this study, literature studies using ISCX-2012 data set, which was widely used in the development of STS systems, were reviewed. Besides, by using this data set, cyber-attacks were detected with 100% accuracy with a powerful machine learning-based approach. Feature and hyperparameter selection algorithms are used to increase the classification accuracy performance of the proposed method. These machine learning approaches are thought to be useful for developing artificial intelligence-based STS systems.

Keywords

• Cyber-attack
• intrusion detection systems
• machine learning

Siber saldırı tespiti için makine öğrenmesi yöntemlerinin performanslarının incelenmesi

Abstract

İnternet tabanlı cihazların kullanımının artması, siber ortamda güvenlik sorunlarına yol açmaktadır. Kötü amaçlı yazılımlar, sistemlerin işleyişini bozabilir ve sistemlerdeki güvenlik açıkları nedeniyle veri gizliliğini tehlikeye atabilir. Siber saldırıları belirlemek ve sınıflandırmak için Saldırı Tespit Sistemleri (STS) geliştirilmektedir. Yapay zeka tabanlı yöntemler, STS sistemlerini iyileştirmek için daha sık kullanılmaktadır. Bu çalışmada, STS sistemlerinin geliştirilmesinde yaygın olarak kullanılan ISCX-2012 veri setinin kullanıldığı literatür çalışmaları gözden geçirilmiştir. Ayrıca bu veri seti kullanılarak makine öğrenmesi tabanlı güçlü bir yaklaşım ile siber saldırılar %100 doğrulukla tespit edilmiştir. Önerilen yöntemin sınıflandırma doğruluğu performansını artırmak için öznitelik ve hiperparametre seçme algoritmaları kullanılmıştır. Önerilen yaklaşımın yapay zeka temelli STS sistemleri geliştirmek için faydalı olacağı düşünülmektedir.

Keywords

• Siber Saldırı
• Saldırı Tespit Sistemleri
• Makine Öğrenmesi

References

1. Kemp, S., Digital 2019: Global digital overview, Retrieved from Datareportal, (2019).
2. Hajisalem, V. ve Babaie, S., A hybrid intrusion detection system based on ABC-AFS algorithm for misuse and anomaly detection, Computer Networks, 136, 37-50, (2018).
3. Inayat, Z., Gani, A., Anuar, N. B., Khan, M. K. ve Anwar, S., Intrusion response systems: Foundations, design, and challenges, Journal of Network and Computer Applications, 62, 53-74, (2016).
4. Ashoor, A. S. ve Gore, S., Difference between intrusion detection system (IDS) and intrusion prevention system (IPS), In International Conference on Network Security and Applications, 497-501, Berlin, Heidelberg, (2011).
5. Jabez, J. ve Muthukumar, B., Intrusion detection system (IDS): anomaly detection using outlier detection approach, Procedia Computer Science, 48, 338-346, (2015).
6. Quepons, I., Vulnerability and Trust, PhaenEx, 13, 2, 1-10, (2020).
7. Garcia-Teodoro, P., Diaz-Verdejo, J., Maciá-Fernández, G. ve Vázquez, E., Anomaly-based network intrusion detection: Techniques, systems and challenges, Computers and Security, 28,1-2, 18-28, (2009).
8. Awad, M., ve Alabdallah, A., Addressing Imbalanced Classes Problem of Intrusion Detection System Using Weighted Extreme Learning Machine, International Journal of Computer Networks & Communications (IJCNC), 1-11, Toronto, Canada, (2019).

9. Bouteraa, I., Derdour, M. ve Ahmim, A., Intrusion Detection using Data Mining: A contemporary comparative study, In 2018 3rd International Conference on Pattern Analysis and Intelligent Systems (PAIS), 1-8, Algeria (2018).
10. Injadat, M., Salo, F., Nassif, A. B., Essex, A. ve Shami, A., Bayesian optimization with machine learning algorithms towards anomaly detection, In 2018 IEEE global communications conference (GLOBECOM), 1-6, Abu Dhabi, (2018).
11. Yassin, W., Udzir, N. I., Muda, Z. ve Sulaiman, M. N., Anomaly-based intrusion detection through k-means clustering and naives bayes classification, In Proc. 4th Int. Conf. Comput. Informatics (ICOCI), 298-303, Kuching, Sarawak, (2013).
12. Hassan, M. M., Gumaei, A., Alsanad, A., Alrubaian, M. ve Fortino, G., A hybrid deep learning model for efficient intrusion detection in big data environment, Information Sciences, 513, 386-396, (2020).
13. Cortes, C. ve Vapnik, V. Support Vector Machine, Machine Learning, 20, 3, 273-297, (1995).
14. Akbulut, Y., Sengur, A. ve Guo, Y., Smarandache, F., NS-k-NN: Neutrosophic set-based k-nearest neighbors classifier, Symmetry, 9, 9, 179, (2017).
15. Safavian, S. R. ve Landgrebe, D., A survey of decision tree classifier methodology, IEEE transactions on systems, man, and cybernetics, 21, 3, 660-674, (1991).
16. Altay, O. ve Ulas, M., Prediction of the autism spectrum disorder diagnosis with linear discriminant analysis classifier and K-nearest neighbor in children, In 2018 6th International Symposium on Digital Forensic and Security (ISDFS), Antalya, Turkey, 1-4, (2018).
17. Kira, K. ve Rendell, L. A., A practical approach to feature selection, In Machine learning proceedings 1992, 249-256, Morgan Kaufmann, (1992).
18. Shiravi, A., Shiravi, H., Tavallaee, M. ve Ghorbani, A. A., Toward developing a systematic approach to generate benchmark datasets for intrusion detection, Computers and Security, 31, 3, 357-374, (2012).
19. Kilincer, I. F., Ertam, F. ve Sengur, A., Machine Learning Methods for Cyber Security Intrusion Detection: Datasets and Comparative Study, Computer Networks, 107840, (2021).
20. Ucar, F. ve Korkmaz, D, COVIDiagnosis-Net: Deep Bayes-SqueezeNet based diagnosis of the coronavirus disease 2019 (COVID-19) from X-ray images, Medical Hypotheses, 140, 109761, (2020).