NEURAL NETWORK FOR MITIGATING DENIAL OF SERVICE (DoS) AND MAN IN THE MIDDLE (MitM) ATTACKS IN CYBERSPACE

Abstract

Denial of Service (DoS) and Man-in-the-Middle (MitM) attacks are serious challenges to network security in the quickly changing world of cyberspace. They disrupt services and jeopardize the confidentiality and integrity of data. Conventional detection techniques frequently find it difficult to keep up with the volume and complexity of contemporary network traffic. Because of their capacity to extract intricate patterns and behaviours from data, Neural Networks (NNs) have become a viable option for intelligent and adaptive intrusion detection in recent years. In order to detect and mitigate DoS and MitM attacks, this review paper examines the use of Neural Networks (NNs). We highlight model performance in terms of accuracy, detection speed, and resilience by presenting comparative insights from recent empirical research. There is also discussion of difficulties including adversarial resilience, model interpretability, and dataset quality. The assessment ends with suggestions for future lines of inquiry meant to improve the efficacy and real-time application of Cyber Security solutions based on neural networks.