A Deep Learning Approach for Intrusion Detection using Recurrent Neural Network
Date
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Department of computer Science & information Technology
Abstract
Intrusion detection discover a critical part in guaranteeing data security and the key
innovation is to precisely recognize different assaults in the system. In this dissertation, the
intrusion detection model based on deep learning is investigated, and a deep learning
approach for intrusion detection using recurrent neural networks (RNN-IDS) is used. The
performance of the model is based on binary and multiclass classification, and the number of
neurons and different learning rate impacts on the performance of the model has been
studied. The performance of the model is compared with Naïve Bayes, Multilayer Perceptron
and Support Vector Machine that has been analyzed by previous researchers on the
benchmark data set. The test results demonstrate that RNN-IDS is remarkably appropriate for
displaying high precision and its execution is better than that of machine learning techniques
in both binary and multiclass classification. The RNN-IDS demonstrate enhances the
precision of the intrusion identification and gives other examination strategy to intrusion
detection discovery.
Keywords: Recurrent neural networks, RNN-IDS, intrusion detection, deep learning,
machine learning.