Intrusion Detection System for NSL-KDD Dataset Based on Deep Learning and Recursive Feature Elimination

Intrusion detection system is responsible for monitoring the systems and detect attacks, whether on (host or a network) identifying attacks that could come to cause damage them, that’s mean an IDS prevents unauthorized access by giving alert administrator before causing any serious harm. As reasonable supplement of firewall, intrusion technology can assist deal with offensive, Intrusions Detection Systems (IDSs) suffers from high false positive which leads highly bad accuracy rate. So this work suggested implement (IDS) using Recursive Feature Elimination select features use Deep Neural Network (DNN) Recurrent (RNN) classification, model gives good results rate reaching 94%, DNN was used in binary classification classify either attack Normal, while RNN classifications five classes (Normal, Dos, Probe, R2L, U2R). The implemented (NSL-KDD) dataset, very efficient offline analyses IDS.