| Machine Learning and Zero-Trust Architectures to Mitigate Blockchain Vulnerabilities and Cryptocurrency Frauds |
| Paper ID : 1063-ICEEM2025 |
| Authors |
|
Konstantinos Lizos *1, Elena Petrovik2, Saied M. Abd El-atty3 1Engineer in the Hellenic Ministry of Foreign Affairs 2Engineer / Independent Network Consultant & Researcher 3The Dept. of Electronics and Electrical Communications Engineering, Faculty of Electronic Engineering, Menoufia University, 32952, Menouf, Egypt |
| Abstract |
| We hereby present the abstract of the submitted paper for the purpose of research evaluation. The expansion of blockchain-based digital currencies has driven significant innovation while simultaneously creating new opportunities for security breaches, financial fraud, and major theft incidents. Although blockchain networks incorporate strong cryptographic protections, attackers consistently find ways to exploit software weaknesses, poor governance structures, and user errors. This study comprehensively examines documented cases of cryptocurrency theft and fraud, evaluates fundamental blockchain security weaknesses, and presents an integrated protection framework combining machine learning algorithms with zero-trust security models. We categorize common attack methods, including Sybil attacks, smart contract vulnerabilities, and advanced social engineering tactics. By applying anomaly detection systems, predictive analytics, and continuous authentication protocols, we show how machine learning-based security solutions can strengthen zero-trust frameworks to counter emerging threats. Our proposed methodology aims to create more robust, adaptive, and verifiable protection systems for blockchain networks and their users. |
| Keywords |
| Machine Learning, Zero-Trust, Blockchain Security, Cryptocurrency Fraud Detection, Anomaly Detection |
| Status: Accepted |