Anomaly Intrusion Detection Design Using Hybrid of Unsupervised and Supervised Neural Network

This paper proposed a new approach to design the system using a hybrid of misuse and anomaly detection for training of normal and attack packets respectively. The utilized method for attack training is the combination of unsupervised and supervised Neural Network (NN) for Intrusion Detection System. By the unsupervised NN based on Self Organizing Map (SOM), attacks will be classified into smaller categories considering their similar features, and then unsupervised NN based on Backpropagation will be used for clustering. By misuse approach known packets would be identified fast and unknown attacks will be able to detect by this method.