LSTM deep learning method for network intrusion detection system

• DOI: 10.11591/ijece.v10i3.pp3315-3322
• Publication Date: 2020-03-08T04:11:12Z
• Authors: Alaeddine Boukhalfa; Abderrahim Abdell...; Nabil Hmina; Habiba Chaoui
• Keywords: Memorization; Identification

The security of the network has become a primary concern for organizations. Attackers use different means to disrupt services or steal information, these various attacks push think new way block them all in one manner. In addition, intrusions can change and penetrate devices security. To solve issues, we suggest, this paper, idea Network Intrusion Detection System (NIDS) based on Long Short-TermMemory (LSTM) recognize menaces obtain long-term memory them, inorder stop that are like existing ones, at sametime, have single mean intrusions. According results experiments detections carried out, Accuracy reaches upto 99.98 % 99.93 respectively classification two classes several classes, Also False Positive Rate (FPR) up only 0,068 0,023 which proves proposed model is very effective, it great ability memorize differentiate between normal traffic attack its identification more accurate than other Machine Learning classifiers.