Building a Novel Intrusion Detection System using Long Short-Term Memory for Detecting Network Attacks

Abstract

With the increased usage of Internet and the advancement of technology the network traffic is heavy and brought major challenge to conventional security mechanisms. Th

e attackers also trying to launch sophisticated attacks to exploit potential vulnerabilities. The traditional intrusion detection system (IDS) is not able to handle massive data. Also, the data available is imbalanced which seriously affect the performance of classifier in IDS. In this paper, sampling technique is used to handle imbalance problem of data set. In addition, a novel Long Short-Term Memory (LSTM) based IDS is introduced to detect the attacks using NSL-KDD data set. The results demonstrates that the proposed system achieve detection accuracy of 92% for binary classification.