An Explainable Machine Learning Framework for AWS S3 CloudTrail-Like Log Analysis in Digital Forensics
International Journal of Advanced Computer Science and Applications · 2026 · 17 (10)
DOI: 10.14569/ijacsa.2026.0171011 ↗Kurz gefasst
Fachartikel in der Zeitschrift International Journal of Advanced Computer Science and Applications, 2026. Verfasst von Aseel Aljuhani und 2 weiteren Autor:innen.
Abstract (Originalsprache)
The advent of cloud computing for data storage and management has led to the emergence of advanced cyber threats that call for the need for effective cloud log analysis for digital forensic investigations. While there has been considerable development of machine learning algorithms for anomaly detection, most of the current approaches focus on detection effectiveness without considering interpretability. This research introduces an explainable hybrid machine learning architecture for analyzing Amazon Web Services (AWS) CloudTrail-like logs in cloud digital forensics. The framework uses XGBoost for classification, Isolation Forest for anomaly detection, rule-based reasoning to detect risky behaviors, and SHapley Additive exPlanations (SHAP) to explain XGBoost predictions. The framework further includes temporal behavior analysis and sensitive cloud object interactions for enhanced cloud digital forensic investigation. The performance of the framework is tested on 169,986 synthetic CloudTrail-like events, with the main evaluation using 35,963 temporally held-out events. The PR-AUC score obtained in the experiments is 0.9324, and ROC-AUC is 0.9827. During monitoring mode, the framework achieves a precision of 0.9677 and recall of 0.7930, whereas forensic mode raises precision to 0.9822 and reduces recall to 0.7844, while keeping the F1-score around 0.87 in both modes. Additional experiments highlight the role of sensitive-object features and demonstrate consistent discrimination across chronological validation folds.
Zitieren
APA
Aljuhani, A., Alhashmi, A. A., & Darem, A. A. (2026). An Explainable Machine Learning Framework for AWS S3 CloudTrail-Like Log Analysis in Digital Forensics. International Journal of Advanced Computer Science and Applications, 17(10). https://doi.org/10.14569/ijacsa.2026.0171011
Vancouver
Aljuhani A, Alhashmi AA, Darem AA. An Explainable Machine Learning Framework for AWS S3 CloudTrail-Like Log Analysis in Digital Forensics. International Journal of Advanced Computer Science and Applications. 2026;17(10). doi:10.14569/ijacsa.2026.0171011
BibTeX
@article{Aljuhani2026,
author = {Aljuhani, Aseel and Alhashmi, Asma A. and Darem, Abdulbasit A.},
title = {An Explainable Machine Learning Framework for AWS S3 CloudTrail-Like Log Analysis in Digital Forensics},
journal = {International Journal of Advanced Computer Science and Applications},
year = {2026},
volume = {17},
number = {10},
doi = {10.14569/ijacsa.2026.0171011}
}
EUVIMED