Enhancing Cloud Security with Machine Learning-Based Intrusion Detection Systems

Pacha, Sai Bhanu Prasad (2025) Enhancing Cloud Security with Machine Learning-Based Intrusion Detection Systems. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (855kB) | Preview
Preview	PDF (Configuration Manual) Download (385kB) | Preview

Abstract

Network intrusion detection systems have severe problems such as high false positive rates, imbalance of classes in training data, and disparity in the laboratory performance and production deployment needs. This study builds and analyses an ensemble machine learning model by integrating the Random Forest and XGBoost model to perform binary classification of network flows based on the CSE-CIC-IDS-2018 dataset. It preprocesses 723,337 samples and standardizes 69 network flow feature using network flow using preprocessing and stratified train-validation-test splitting. The accuracy of individual models is 87.7% (Random Forest) and 87.6% (XGBoost), whereas the weighted ensemble exhibits a high accuracy of 87.84% and a high precision and recall value of 95.2% and 78.3% respectively. The system runs on AWS EC2 through Docker containerization, where the median inference latency of the Flask REST API is 376ms and median server-side processing time is 42ms. Extensive monitoring solutions based on Prometheus and Grafana help track the performance in real-time, and a set of custom analytics solutions allow analysis of the latency distribution, evaluate its accuracy through confidence, and suggest the optimal threshold. The deployment has been shown to be 100% available during hours of testing with steady performance characteristics, indicating that it is ready to be deployed to production to provide real-time network intrusion detection with balanced accuracy and performance.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Samarawickrama, Yasantha 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 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:	01 Sep 2026 09:33
Last Modified:	01 Sep 2026 09:33
URI:	https://norma.ncirl.ie/id/eprint/9725

Actions (login required)

View Item