CloudShield AI: An Explainable, Isolation-Forest-Based Prototype for Detecting Suspicious Data Access in Cloud Activity Logs
This research presents CloudShield AI, a hybrid framework designed to identify suspicious user activities and potential data theft in cloud environments. The proposed approach combines machine learning-based anomaly detection using Isolation Forest with rule-based risk analysis to examine user activity logs. The system performs data preprocessing, duplicate removal, suspicious activity identification, risk classification, and generates explanations for detected anomalies. It also incorporates AES-based encryption to protect exported analysis reports. The framework is evaluated using a sample dataset to demonstrate its workflow and functionality. The study discusses its potential applications, limitations, and future improvements for cloud security monitoring.
Authors
- Dhruv Jadhav (ORCID: https://orcid.org/0009-0008-8684-4568)
- MHAMANE SHIVRAJ
- BHADANE SHUBHAM
- NAVALE YASHRAJ
Publication Details
- Journal
- Zenodo (CERN European Organization for Nuclear Research)
- Published
- 2026-11-17
- DOI
- https://doi.org/10.5281/zenodo.24325609
- Primary Topic
- Cloud Data Security Solutions
- Type
- article
- Field-Weighted Citation Impact
- 0.00