An Intrusion Detection Model Based on Deep Long Short Term Recurrent Neural Network
K. Narayana Rao1, K. Venkata Rao2, Prasad Reddy PVGD3
1K. Narayana Rao*,Research Scholar, Department of Computer Science and Systems Engineering, Andhra University College of Engineering (A), Andhra University, Visakhapatnam, AP, India.
2Prof. K. Venkata Rao, Professor, Department of Computer Science and Systems Engineering, Andhra University College of Engineering (A), and DeanAcademic Affairs, Andhra University, Visakhapatnam, AP, India.
3Prof. Prasad Reddy P.V.G.D, Sr. Professor, Department of Computer Science and Systems Engineering, Andhra University College of Engineering (A), and Vice-Chancellor, Andhra University, Visakhapatnam, AP, India.
Manuscript received on November 22, 2019. | Revised Manuscript received on December 15, 2019. | Manuscript published on December 30, 2019. | PP: 2870-2875 | Volume-9 Issue-2, December, 2019. | Retrieval Number: B3640129219/2019©BEIESP | DOI: 10.35940/ijeat.B3640.129219
Open Access | Ethics and Policies | Cite | Mendeley
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: With the rapid increase of network based services and internet users on various platforms are becoming the major targets of attacks. Intrusion detection is the process of monitoring the attacks and analyzing their signs and violation of security policies which are occurring in the systems or networks. Intrusion Detection System is a prominent research area in security analysis and evaluation. In order to identify the attack type, we proposed Deep Long Short Term Memory-Recurrent Neural Network (DLSTM-RNN) method with seven optimizers and 500 epochs to train and test a dataset. Initially the data transformation, normalization are used to preprocess the data. The preprocessed train and test data is given input to the model. The bench mark NSL-KDD dataset used to train and test the model. The results are obtained for five-class classification (attack types).The model outperformed with adamax optimizer on NSL-KDD dataset. The metrics accuracy, detection rate, and false alarm rate areevaluated to ascertain the detection efficacy of the model. We compare the model to existing convolutional learning methods.
Keywords: Deep Learning, Long Short Term Memory, Optimizer, Intrusion Detection.