Privacy-Preserving Cybersecurity in Cloud Computing Environments Using Artificial Intelligence-Based-Classification Model

Abstract

Cloud computing (CC) has revolutionized data management, but it continues to face critical cybersecurity challenges, particularly in preserving privacy and detecting threats. This study presents a novel AI-driven framework that integrates feature selection, neuro-fuzzy classification, adaptive encryption, and metaheuristic optimization to enhance privacy-preserving cybersecurity in cloud environments. The proposed methodology uses the term frequency-inverse document frequency (TF-IDF) for dimensionality reduction, an enhanced adaptive neuro-fuzzy inference system (ANFIS) for attack detection, an advanced cryptographic standard technique (ACST) for secure encryption, and the Archimedes Optimization Algorithm (AOA) for hyperparameter tuning. Experimental results demonstrate improved classification accuracy over conventional methods, efficient and robust encryption, and optimized performance suitable for real-time deployment. The framework strikes a balance between detection accuracy and computational efficiency while ensuring compliance with regulatory requirements, such as Indonesia’s data sovereignty laws. These findings suggest that integrating adaptive AI techniques with lightweight cryptography offers a scalable and effective approach to cloud security. Practical implications include enhanced protection of sensitive data in multi-tenant environments and alignment with evolving data protection regulations. Future research should explore quantum-resistant encryption and federated learning (FL) to strengthen cross-cloud collaboration and resilience.