Abstract

Network intrusion attacks have significantly increased in recent years, which creates serious privacy and security concerns. Technology development has made cyber-security threats more advanced, to the point where the detection mechanisms in place are unable to handle the problem. Therefore, the key to solving this issue would be the deployment of a reliable Network Intrusion Detection System (NIDS). In this work, an intrusion detection system is developed that can identify various network attacks using deep learning techniques, including Graph Neural Network (GNN) and Recurrent Neural Network (RNN).By comparing and evaluating the performance of the presented results using various evaluation metrices, the most effective framework for the network intrusion detection system could be discovered.