Intrusion Detection using an Improved Competitive Learning Lamstar Neural Network

Computer systems vulnerabilities such as software bugs are often exploited by malicious users to intrude into information systems. With the recent growth of the Internet such security limitations are becoming more and more pressing. One commonly used defense measure against such malicious attacks in the Internet are Intrusion Detection Systems (IDSs). Due to increasing incidents of cyber attacks, building effective intrusion detection systems (IDS) are essential for protecting information systems security, and yet it remains an elusive goal and a great challenge. We developed an Intrusion Detection System using LAMSTAR neural network to learn patterns of normal and intrusive activities and to classify observed system activities. we further investigate the time taken for training and testing, generate Confusion matrix with KDD CUP 99 data using simulation tool and compare it with five classification techniques (Gaussian Mixture, Radial Basis Function, Binary Tree Classifier, SOM, and ART ). The results indicate that LAMSTAR exhibit high accuracy at the cost of long training time.