Adaptive Detection of Advanced Persistent Threats (APT) in Multi-Layered Network Environments

Guttikonda, Bhaskar (2024) Adaptive Detection of Advanced Persistent Threats (APT) in Multi-Layered Network Environments. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (762kB) | Preview
Preview	PDF (Configuration Manual) Download (383kB) | Preview

Abstract

Advanced Persistent Threats (APTs) represent an increasingly sophisticated challenge in cybersecurity, requiring innovative approaches for effective detection and mitigation. This research presents a comprehensive multi-layered detection framework that integrates and machine learning techniques including deep learning, natural language processing to identify and analyze APT activities across different attack vectors. The implementation combines network traffic analysis, phishing URL detection, and keylogger activity monitoring, achieving significant detection accuracy across all components. The network traffic analysis module, using a Sequential Neural Network architecture, achieved an accuracy of 99.30% in classifying different attack patterns. The phishing detection module showed the accuracy of 96.45% by using the combined methods of NLP and machine learning, while the keylogger detection system achieved an accuracy of 96% using tree-based models. Feature importance analysis presented important patterns across the attack vectors; flow-based metrics and behavioral patterns were critical indicators. It provides real-time correlation capabilities and adaptive response mechanisms that might cure serious shortcomings of the current APT detection approaches. Though computational overhead and integration costs remain a challenge, the proposed framework has great potential for real-world enterprise deployment. The results add to the theoretical understanding of integrated APT detection system design and to the development of practical implementations, providing a base for future research in adaptive security mechanisms and real-time threat detection.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Khan, Imran UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science P Language and Literature > P Philology. Linguistics > Computational linguistics. Natural language processing Q Science > QA Mathematics > Computer software > Computer Security T Technology > T Technology (General) > Information Technology > Computer software > Computer Security Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Cyber Security
Depositing User:	Ciara O'Brien
Date Deposited:	23 Jul 2025 13:42
Last Modified:	23 Jul 2025 13:42
URI:	https://norma.ncirl.ie/id/eprint/8212

Actions (login required)

View Item