Boost-AGL Stack: A Scalable and Secure Ensemble Approach for Malicious URL Detection in the Cloud

Dasari, Lakshmi Narasimha Naga Manikanta (2025) Boost-AGL Stack: A Scalable and Secure Ensemble Approach for Malicious URL Detection in the Cloud. Masters thesis, Dublin, National College of Ireland.

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

Abstract

Malicious URL detection is a critical task in cybersecurity, particularly as phishing attacks increasingly exploit cloud-hosted platforms and services. Traditional approaches often rely on static blacklisting or standalone machine learning models, which struggle with adaptability, scalability, and real-time responsiveness—key requirements in dynamic cloud environments. These methods typically lack deployment automation, consistent performance under varying loads, and integration with modern MLOps workflows. To address these limitations, this study proposes a cloud-native, scalable solution named Boost-AGL Stack, a stacking ensemble model that combines AdaBoost and Gradient Boosting as base learners with Logistic Regression as a meta-classifier. The model is deployed using Docker containers on AWS EC2, with CI/CD automation handled via GitHub Actions, ensuring seamless updates and reliable cloud operation. Experiments conducted on the ISCX-URL-2016 dataset demonstrate that the Boost-AGL Stack outperforms traditional models, achieving the highest accuracy 91%, compared to baseline models like Logistic Regression (90%) and AdaBoost (85%). This confirms the ensemble’s robustness and generalization ability. By integrating feature-based classification with cloud deployment strategies, the proposed approach delivers a secure, scalable, and production-ready solution for real-time phishing URL detection. The findings highlight its potential for modern cybersecurity systems operating in cloud ecosystems.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Samarawickrama, Yasantha UNSPECIFIED
Uncontrolled Keywords:	Cloud Computing; Malicious URL Detection; Stacking Ensemble; MLOps; AWS Deployment
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:	20 Mar 2026 15:00
Last Modified:	20 Mar 2026 15:00
URI:	https://norma.ncirl.ie/id/eprint/9207

Actions (login required)

View Item