AI-Driven Self-Learning Intrusion Detection & Response System for Cloud Security

Lahane, Rohan Vivek (2025) AI-Driven Self-Learning Intrusion Detection & Response System for Cloud Security. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (887kB) | Preview

Abstract

Sophisticated cyber threats in cloud environments, especially those that are hosted on platforms such as Amazon Web Services (AWS) are shedding more and more since they are complex, and scalable in nature as well as their dynamic user access model. The current types of attacks on clouds that exploit identity and access management misconfigurations are quite difficult to be tracked with traditional intrusion detection systems (IDS). The present project suggests an AI-based, self-educating intrusion Detection and Response System (IDRS) that uses both supervised and unsupervised machine learning (XGBoost, Isolation Forest) in detecting abnormalities in the behaviour of cloud traffic.

Locally, the system is prototyped with cleaned KDD datasets and simulated AWS-like CloudTrail logs to mimic the real-world clouds, and their activities with unauthorized access, privileged escalation, and reconnaissance attempts. The solution has proven the power of using both AI models and simulated identity-driven detection of threats to secure AWS-style environment in the case that there is no direct cloud connection. The results of the model are improved through the measurement of the performance of the model in terms of accuracy, precision, recall, and confusion matrices.

This paper establishes a fault-tolerant future deployment of AWS by suggesting how it is possible to translate this system into consuming genuine CloudTrail logs, generating automatic response, and interfacing with SOAR solutions. Although locally developed, the architecture is entirely scalable to a cloud-native environment, and opens the door to automating response and continuous learning in contemporary cyber defence.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Pantridge, Michael UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence T Technology > T Technology (General) > Information Technology > Cloud computing Q Science > QA Mathematics > Computer software > Computer Security T Technology > T Technology (General) > Information Technology > Computer software > Computer Security
Divisions:	School of Computing > Master of Science in Cyber Security
Depositing User:	Ciara O'Brien
Date Deposited:	18 Aug 2026 16:39
Last Modified:	18 Aug 2026 16:39
URI:	https://norma.ncirl.ie/id/eprint/9538

Actions (login required)

View Item