A New Approach for Detecting Intrusions Using Jordan/Elman Neural Networks

Abstract

Intrusion detection system (IDS) is an effective tool that can help to prevent unauthorized access to network resources. A good intrusion detection system should have higher detection rate and lower false positive. A new classification system using Jordan/Elman (J/L) neural network for ID is proposed to detect intrusions from normal connections with satisfactory detection rate and false positive. Experiments and evaluations were performed with the KDD Cup 99 intrusion detection database. This system yields the same performance level or better as compared to other existing systems. Comparison with other approach based on different evaluation parameters showed that proposed approach has noticeable performance with detection rate 99.594% and false positive 0.406% and can classify the network connections with satisfactory performance.