Enhancing Cloud Access Control: Leveraging Machine Learning for Security Score Prediction and Improvement

Singh, Anurag (2024) Enhancing Cloud Access Control: Leveraging Machine Learning for Security Score Prediction and Improvement. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (2MB) | Preview

Abstract

Flexibility and scale have made cloud services essential for modern organizations, still, big security challenges surrounding access control management come with that. Role Based Access Control (RBAC) and Attribute Based Access Control (ABAC) serve as the fundamentals, however, existing methods are insufficient to describe the dynamic nature of the cloud from the perspective of cloud access control enforcement that changes along with it. In recent years, machine learning has been employed to improve cloud access control via anomaly detection, policy mining, and predictive analysis. Despite this, most previous solutions are limited to detection and do not integrate predictive solutions with actionable feedback for real-time remediation. To address these limitations, this research leverages predictive analytics alongside automated feedback to proactively remediate cloud access control by closing the accurate prediction and remediation gap. Applying the ML models to make security score predictions and feedback in access control configuration. XGBoost is the most robust model due to optimization across hyperparameters of RandomizedSearchCV. Through a novel feedback mechanism, vulnerabilities are automatically identified based on predictions. It consists of a simple yet functional user interface and Flask-based REST API which gives real-time insights and actionable recommendations. The solution is deployed on top of AWS services (Cloud9, S3, Elastic Beanstalk, CodePipeline) for scalability and convenient Integration. The results show proactive security management with reduced risk exposure and better compliance. Through the combination of predictive analytics with dynamic feedback, this research pushed the boundaries of cloud security and demonstrated that adaptive learning and cloud interoperability will be necessary for future development.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Kazmi, Aqeel UNSPECIFIED
Uncontrolled Keywords:	Access Control; Machine Learning; Cloud9; S3; Elastic Beanstalk; XGBoost; Flask Framework; Hyperparameter Tuning; Auto-Remediation; CodePipeline; Compliance
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science 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:	16 Jul 2025 14:07
Last Modified:	16 Jul 2025 14:07
URI:	https://norma.ncirl.ie/id/eprint/8156

Actions (login required)

View Item