A Hybrid AI-Based Intrusion Detection System Using Ensemble Learning and Deep Packet Inspection for Encrypted Network Traffic
Keywords:
Intrusion Detection System, Hybrid AI, Ensemble Learning, Deep Packet Inspection, Encrypted Network Traffic, Machine Learning, Classification, Network Security, Cybersecurity, Anomaly DetectionAbstract
The increasing volume of encrypted network traffic poses a significant challenge to traditional Intrusion Detection Systems (IDS). This paper proposes a hybrid approach combining ensemble learning techniques with Deep Packet Inspection (DPI) to detect intrusions in encrypted network traffic. The proposed system uses machine learning algorithms for classification and ensemble techniques to enhance the detection accuracy. DPI is applied to analyze encrypted packets, identifying malicious patterns without needing to decrypt the data. The effectiveness of the hybrid system is evaluated using various benchmark datasets, demonstrating its ability to detect a wide range of network attacks with high accuracy, precision, and recall. The results show that ensemble learning models outperform traditional IDS, providing a robust solution for securing encrypted communications in modern networks.
References
Zhang, X., Liu, Y., & Li, Z. (2020). Intrusion detection systems and encrypted traffic analysis. Springer.
Muhammad, S., & Arshad, M. (2019). Machine learning techniques for improving intrusion detection systems. Springer.
Kamadi, S. (2023). Identity-Driven Zero Trust Automation in GitOps: Policy-as-Code Enforcement for Secure Code Deployments. International Journal of Scientific Research in Computer Science, Engineering and Information Technology, 9(3), 893-902. https://doi.org/10.32628/CSEIT235148
Pham, L., & Ha, T. (2021). Deep packet inspection for encrypted traffic anomaly detection. Springer.
Xu, J. (2020). Enhancing IDS performance with ensemble learning: A study of random forests and gradient boosting. Springer.
Kumar, S., & Bhadoria, R. (2018). Hybrid machine learning and DPI for intrusion detection systems. Springer.
Silva, C., Costa, J., & Oliveira, L. (2021). Neural network integration with deep packet inspection for detecting malicious traffic. Springer.
Diaz, F. (2019). Using AdaBoost to reduce false positives in network intrusion detection systems. Springer.
Liu, Q., Zhang, H., & Sun, X. (2020). Ensemble learning for handling class imbalance in network intrusion detection. Springer.
Patel, K., & Raghunathan, R. (2020). Securing modern networks with hybrid DPI and machine learning models. Springer.
Zhang, L., Wang, S., & Zhang, Z. (2019). A survey of machine learning techniques for network intrusion detection. Journal of Network and Computer Applications, 123, 1-14.
Wang, H., & Liu, Q. (2021). Deep learning for encrypted traffic detection: A review and future directions. Computers & Security, 95, 101892.
Xu, B., & Wu, Y. (2018). An efficient ensemble learning approach for intrusion detection in large-scale networks. Information Sciences, 444, 334-345.
Liu, Y., & Yang, S. (2020). Traffic analysis with deep packet inspection and machine learning for enhanced IDS performance. Computer Networks, 178, 107322.
Ahmad, S., & Gupta, M. (2021). A hybrid deep learning model for anomaly-based network intrusion detection. Security and Privacy, 4(3), e167.
