Intrusion Detection System using Deep Learning
The security of the network is of great concern as the network space is facing constant attack by intruder and they are becoming Spontaneous and harder to curtail. Many hardware, software and models created and deployed have shown great results in detection and incident response rates, but more work needs to be done, as many malicious attacks with more advance attack vectors can easily get past most Network security mechanisms undetected and compromise security features like confidentiality, integrity, availability etc. In this work, an Intrusion Detection System Model was designed and implemented based on feature extracted from CIC-IDS 2017 dataset using Convolutional Neural Network (CNN). Feature learning and modeling were achieved by analyzing the Attack types in the Dataset. The proposed system was implemented in Python and performance is measured using accuracy, precision and recall. The results obtained from the implementation of the CNN-IDS model’s prediction, accuracy evaluation, when compared to other Models proved that the Convolutional Neural Network model approach to Intrusion Detection performed better in accuracy 99.78%.
Authors: Ayeni Olaniyi A., Ewa Stanley C., Owolafe Otasowie
- Date of Conference: 6-8 December 2022
- DOI: 10.20533/ICITST.2022.0005
- ISBN: 978-1-913572-55-6
- Conference Location: Virtual (London, UK)