ML-Based Data Leakage and Tampering Detection System in Enterprise Data Workflows
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
Modern enterprises face escalating risks of insider threats, data leakage, and document tampering as digital workflows grow increasingly complex.Conventional rule-based security tools fail to detect adaptive, authorized-user attacks that exploit behavioral and semantic anomalies.This study presents a comprehensive survey and conceptual framework integrating behavioral analytics, automated content categorization, and document integrity verification using advanced machine learning models such as Isolation Forests for anomaly detection and transformer-based language models for semantic analysis.The proposed multi-modal framework emphasizes transparency, explainability, and adaptability across enterprise environments.A simulated evaluation using synthetic enterprise datasets demonstrates detection accuracy of 96.2 %, a false positive rate of 2.0 %, and scalability across varied workloadsshowing the framework's feasibility for real-time, ML-driven data protection.The paper also identifies key research gaps and outlines future directions toward privacy-preserving, federated, and explainable AI-based enterprise security architectures.
Fields of Study & Subjects
Privacy-Preserving Technologies in Data
Network Security and Intrusion Detection
Information and Cyber Security
Publication details
- Journal / source
- Advances in intelligent systems research/Advances in Intelligent Systems Research
- Pages
- 81–96
- Publication date
- 1 April 2026
- Language
- EN
- Document type
- Book chapter