The complexity of emerging cyber threats renders the need for intelligent, adaptive and real-time responses essential. In this paper, an explainable deep learning assisted federated learning and adversarial robustness-based AI-enabled threat detection scheme for cybersecurity is introduced. The framework also breaks the limitations of traditional models with continuous learning, cross-platform fusion, proactive exception detection. By using lightweight neural architectures and fusing threat information, the system achieves scale and low latency. Experiments demonstrate a 96.8% attack detection rate, and 42% decrease in false alarm rate outperforming the state-of-the-art.
K. Jamberi, S. Rajeshwari, G. Prasadu et al.· 0 citations
Edge computing is developed as a promising solution operating and performing modern web applications with low latency, fast load times, and efficient resource uses near the source of the data. This content describes an end-to-end framework involving front-end, back-end, and edge-layer optimizations, while introducing an empirical-based benchmark, as well as a security integration, and compatibility with different frameworks, to overcome the limitations of current approaches. We have also implemented the proposed approach in real-time on various platforms, and have shown large improvements in web response time, and scalability. This work helps to improve edge-native web architecture for performance-sensitive applications.
S. Kannadhasan, Gopinath Anjinappa, Akshay Kumar et al.· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
