Intrusion Detection Using an Ensemble Deep Learning Approach

Abstract

Today's cyber society faces a serious intrusion detection security issue. Recent years have seen a sharp rise in network intrusion attacks, raising severe privacy and security concerns. The complexity of cyber-security threats is increasing due to technological improvement, making it impossible for the current detection methods to handle the problem. So, creating an intelligent and efficient network intrusion detection system would be crucial to resolving this problem. In this paper, we created an intelligent intrusion detection system that can detect different networking attacks using deep learning approaches, specifically Convolutional Neural Networks (CNN) and Deep Neural Networks (DNN). We used an ensemble model of CNN and DNN which provides us with great accuracy. Before being used for model training and testing, the obtained data is analysed and pre-processed. Also, in order to choose the optimum model for the network intrusion detection system, we compared the outcomes of our proposed solution and evaluated the performance of the proposed solution using several evaluation matrices.