Threat Detection and Intrusion Prevention in Cloud-Based Infrastructures

Joshy, Abhishek (2025) Threat Detection and Intrusion Prevention in Cloud-Based Infrastructures. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (343kB) | Preview
Preview	PDF (Configuration Manual) Download (133kB) | Preview

Abstract

The rapid migration of enterprise infrastructure to cloud computing environments has introduced new and complex security challenges. Traditional security perimeters are dissolving, necessitating advanced, intelligent systems capable of monitoring vast and dynamic network traffic for malicious activities. This research addresses the critical need for effective threat detection and intrusion prevention within cloud-based infrastructures. The project develops and evaluates a comprehensive solution by leveraging a large-scale public dataset, CIC-IDS2017, to train a suite of machine learning models. A systematic methodology involving data preprocessing, feature engineering, and dimensionality reduction was employed to prepare the data. Multiple classification algorithms were trained and rigorously evaluated, with a Decision Tree model emerging as the optimal choice, achieving a classification accuracy of 97.4%. The core contribution of this work is the operationalization of this high-performance model within a cloud-native architecture. A Flask web application was developed to serve as the system's engine, featuring a real-time analytics dashboard and a background process for continuous, simulated packet scanning. This entire system was deployed on Amazon Web Services (AWS), demonstrating a practical, end-to-end implementation. A key innovation is the deep integration with AWS CloudWatch, enabling the system to export custom security metrics and logs for centralized monitoring, alerting, and long-term analysis. The final artifact is not merely a theoretical model but a fully functional prototype that provides a blueprint for deploying intelligent, scalable, and resilient intrusion detection systems in modern cloud environments.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Heeney, Sean UNSPECIFIED
Subjects:	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 T Technology > T Technology (General) > Information Technology Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Cloud Computing
Depositing User:	Ciara O'Brien
Date Deposited:	26 Mar 2026 10:11
Last Modified:	26 Mar 2026 10:11
URI:	https://norma.ncirl.ie/id/eprint/9221

Actions (login required)

View Item