CYBER THREAT DETECTION AND MITIGATION IN CLOUD INFRASTRUCTURE THROUGH DEEP NEURAL NETWORKS

Abstract

This research employs deep neural networks to detect and mitigate cyber threats to cloud infrastructure in order to address the heightened complexity and diversity of contemporary intrusions. Cloud infrastructures are susceptible to data breaches, insider threats, and DDoS attacks due to their dynamic resource allocation and multi-tenant architectures. Zero-day and intricate attacks are overlooked by conventional rule- and signature-based security. In order to circumvent these constraints, deep neural networks identify intricate patterns and anomalies in extensive cloud traffic and system data. False positives are diminished and fraudulent activities are identified in real time through multi-dataset training. Access control, traffic filtering, and dynamic resource isolation comprise automatic threat mitigation. Experimental results indicate that the DNN-based system outperforms conventional methods in terms of detection accuracy, scalability, and response time, rendering it a reliable cloud infrastructure protection solution against the proliferation of cyber threats