Enhancing Network Security: A Comprehensive Analysis of Intrusion Detection Systems
Öz
Given the increasing complexity and progress of intrusion attacks, effective intrusion detection systems have become crucial to protecting networks. Machine learning methods have become a potential strategy for identifying and reducing such attacks. This paper has conducted a comprehensive analysis of intrusion detection using machine learning methodologies. The aim is to thoroughly examine the current state of research, identify the barriers, and highlight potential solutions in this field. The study begins by analyzing the importance of intrusion detection and the limitations of traditional rule-based systems. Afterward, it explores the underlying principles and concepts of machine learning and how they are practically applied in the field of intrusion detection. This paper provides a comprehensive analysis of different machine learning algorithms, such as decision trees, neural networks, support vector machines, and ensemble methods. The primary objective of this study is to assess the effectiveness and limitations of employing these techniques for identifying various forms of intrusions. Three algorithms are used to classify the NSL-KDD dataset, namely Cascade Backpropagation Neural Networks (CBPNN), Layered Recurrent Neural Networks (LRNN), and Forward-Backward Propagation Neural Networks (FBPNN). Results have shown that CBPNN outperformed by achieving 95% accuracy.
Anahtar Kelimeler
CBPNN, Cyber security, FBPNN, Intrusion detection systems (IDS), Logistic regression, Machine learning
Kaynakça
- Alazab, M., Venkatraman, S., Watters, P., & Alazab, M. (2011). Zero-day malware detection based on supervised learning algorithms of API call signatures. AusDM, 11, 171-182.
- Avcı, İ., & Koca, M. (2023). Cybersecurity attack detection model, using machine learning techniques. Acta Polytechnica Hungarica, 20(7), 2023–2052.
- Bahlali, A. R., & Bachir, A. (2023). Machine learning anomaly-based network ıntrusion detection: experimental evaluation. Lecture Notes in Networks and Systems, 654 LNNS, 392–403. https://doi.org/10.1007/978-3-031-28451-9_34
- Bengio, Y., Simard, P., & Frasconi, P. (1994). Learning long-term dependencies with gradient descent is difficult. IEEE Transactions on Neural Networks, 5(2), 157-166. https://doi.org/10.1109/72.279181
- Biermann, E., Cloete, E., & Venter, L. M. (2001). A comparison of intrusion detection systems. Computers & Security, 20(8), 676–683. https://doi.org/10.1016/S0167-4048(01)00806-9
- Can, O., & Sahingoz, O. K. (2015). A survey of intrusion detection systems in wireless sensor networks. 6th International Conference on Modeling, Simulation, and Applied Optimization, ICMSAO 2015 - Dedicated to the Memory of Late Ibrahim El-Sadek. https://doi.org/10.1109/ICMSAO.2015.7152200
- Çakmak, M., Albayrak, Z., & Torun, C. (2021). Performance comparison of queue management algorithms in LTE networks using NS-3 simulator. Technical Gazette, 28(1), 135-142. https://doi.org/10.17559/TV-20200411071703
- Eskin, E., Arnold, A., Prerau, M., Portnoy, L., & Stolfo, S. (2002). A geometric framework for unsupervised anomaly detection.In Barbará, D., Jajodia, S. (Eds). Applications of data mining in computer security. Advances in Information Security, vol 6. (pp. 77–101). Springer, Boston. https://doi.org/10.1007/978-1-4615-0953-0_4
- Gao, Y., Li, X., Peng, H., Fang, B., & Philip, S. Y. (2020). Hincti: A cyber threat intelligence modeling and identification system based on heterogeneous information network. IEEE Transactions on Knowledge and Data Engineering, 34(2), 708–722. https://doi.org/10.1109/TKDE.2020.2987019
- García-Teodoro, P., Díaz-Verdejo, J., Maciá-Fernández, G., & Vázquez, E. (2009). Anomaly-based network intrusion detection: Techniques, systems and challenges. Computers & Security, 28(1–2), 18–28. https://doi.org/10.1016/J.COSE.2008.08.003