HYBRID INTRUSION DETECTION SYSTEM COMBINING OF SELF-ORGANIZING MAP AND BACKPROPAGATION WITH GANS NEURAL NETWORKS
The increasing complexity of network cyber-attacks has made intrusion detection systems (IDS) a vital component of network security. This paper proposes a hybrid IDS that combines Self-Organizing Map (SOM) and Back propagation neural networks with Generative Adversarial Networks (GANs) for improved network security. By utilizing the SOM and Back propagation neural network, the traffic patterns can be classified and anomalies can be detected. The detection system’s accuracy is improved by training it with synthetic data generated by GANs. The expected results demonstrate that the hybrid system achieves higher accuracy and detection rates compared to using each individual component alone. GANs enable the system to learn and adapt to new attack patterns, making it a robust and effective tool for enhancing network security. The proposed system offers a hopeful method for the timely detection and prevention of potential network intrusions.
Self-Organization Map; Back Propagation Artificial Neural Networks; Intrusion Detection; Generative Adversarial Networks.