Title: Development of IDS using mining and machine learning techniques to estimate DoS malware

Authors: G. Revathy; P. Sathish Kumar; Velayutham Rajendran

Addresses: Department of Electronics and Communication Engineering, Vels Institute of Science, Technology & Advanced Studies (VISTAS), Pallavaram, Chennai, 600117, India ' Department of Electronics and Communication Engineering, Vels Institute of Science, Technology & Advanced Studies (VISTAS), Pallavaram, Chennai, 600117, India ' Department of Electronics and Communication Engineering, Vels Institute of Science, Technology & Advanced Studies (VISTAS), Pallavaram, Chennai, 600117, India

Abstract: A denial of service is a main type of cyber security attack. Intrusion detection system techniques play a very important role for detecting and preventing mechanisms that eradicate the issues made by hackers in the network environment. In this research, we describe different data mining techniques which can be used to handle different kinds of network attacks. Three machine learning techniques are used for classification problems, such as decision tree classifier, gradient boosting classifier, K-nearest neighbour classifier, to find the metric values of false negative rate, accuracy, F-score and prediction time. We found that the decision tree classifier and voting classifier is the best method which has less prediction time and better accuracy of 99.86% and 99.9% which makes the model better along with greater performance. The result shows high accuracy level and less prediction time. Moreover, the relationships between existing approach and proposed approaches in terms of metrics are described.

Keywords: denial of service; DoS; machine learning techniques; MLT; statistical analysis; false negative rate; FNR; intrusion detection system; IDS.

DOI: 10.1504/IJCSE.2021.115646

International Journal of Computational Science and Engineering, 2021 Vol.24 No.3, pp.259 - 275

Received: 13 May 2020
Accepted: 12 Oct 2020

Published online: 04 Jun 2021 *

Full-text access for editors Access for subscribers Purchase this article Comment on this article