A Comprehensive Review of ML-Based Cloud Security: Biometric Authentication, Threat Detection, and Malware Analysis
Abstract: A comprehensive review of machine learning (ML) techniques for enhancing cloud security through biometric authentication, threat detection, and malware analysis. It explores how advanced ML and deep learning models—such as CNNs, RNNs, BiLSTM, SVMs, and Random Forest—are applied to safeguard cloud environments. The review categorizes over 60 recent studies, highlighting key methods for intrusion detection, malware classification, and anomaly detection. Special attention is given to multimodal biometric authentication using ECG, fingerprint, and facial recognition, as well as the integration of adaptive encryption and multi-factor authentication (MFA) for robust access control. Practical use cases in healthcare, IoT, and secure remote services are discussed, alongside emerging trends such as Zero Trust Architecture and self-learning AI. By synthesizing state-of-the-art research, this study provides a layered framework for developing intelligent, resilient, and privacy-preserving cloud security systems.