Deep Learning for Cyberthreats: Performance Analysis and Application of Malware Classification in Edge Computing

In the context of the escalating malware threat landscape and the ongoing challenge of identifying modern malware, this research investigates the utility of Deep Neural Networks (DNNs) for malware classification. Various DNN architectures are explored, revealing their ability to accurately classify malware across different types with consistently high accuracy rates. The study also examines the feasibility of deploying DNN models on edge devices for real-time classification in resource-constrained settings, emphasizing the importance of efficient performance optimization. This research contributes to advancing malware detection techniques and highlights the significance of early malware detection for preventing adverse consequences. It additionally addresses the distribution of security tasks to edge devices to maintain the integrity and availability of large-scale IoT systems, fostering a more resilient and secure digital ecosystem.