Unsupervised learning for Anomaly and Misconfiguration detection of cloud storage services

Maseno, Gift Neema (2025) Unsupervised learning for Anomaly and Misconfiguration detection of cloud storage services. Masters thesis, Dublin, National College of Ireland.

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

Abstract

Lately, there has been an increasing number of cyber threats arising from cloud service misconfigurations as more organisations adopt cloud technologies. Previous rule-based solutions often fail to adapt to the dynamic nature of cloud environments and noisy data, whilst cloud-native solutions can be challenging to understand and implement, leaving S3 buckets susceptible to various threats such as unauthorised access. With the rise of AI & ML implementations in automated anomaly detection, this study explores the feasibility of applying unsupervised deep learning techniques in identifying misconfigured S3 buckets. This study utilised an open-source dataset from the flAWS challenge environment by Summit Route. Since the dataset was highly imbalanced, a rule-based transformation of data was adopted to simulate both secure and misconfigured fields. The autoencoder model was then trained on secure features to learn normal patterns and detect anomalies via reconstruction error. The model achieved commendable results with an accuracy of 89.6%, precision of 99.7%, recall of 76.9% and f1-score of 0.868, showing the model’s capability to detect anomalies with minimal false positives. These results demonstrate the potential of unsupervised models in efficiently detecting and responding to anomalies in cloud environments.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Mustafa, Raza Ul UNSPECIFIED
Uncontrolled Keywords:	Cloud security; S3 buckets; Misconfigurations; Unsupervised learning
Subjects:	Q Science > QH Natural history > QH301 Biology > Methods of research. Technique. Experimental biology > Data processing. Bioinformatics > Artificial intelligence Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Artificial intelligence 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 Cyber Security
Depositing User:	Ciara O'Brien
Date Deposited:	03 Sep 2026 11:34
Last Modified:	03 Sep 2026 11:34
URI:	https://norma.ncirl.ie/id/eprint/9805

Actions (login required)

View Item