Title: An intelligent neuro-genetic framework for effective intrusion detection

Authors: K.P. Rama Prabha; N. Jeyanthi

Addresses: School of Information Technology and Engineering, VIT University, Vellore, India ' School of Information Technology and Engineering, VIT University, Vellore, India

Abstract: In this paper, a new intelligent neuro-genetic framework is proposed for detecting the intruders in networks by analysing their behaviour. For this purpose, a new genetic algorithm based feature selection algorithm (GAFSA) and a neuro-genetic fuzzy classification algorithm (NGFCA) have been proposed in this paper which are used to identify the malicious users through classification of user behaviours. The main advantage of this proposed framework is that it reduces the attacks by identifying the intruders with high accuracy and reduced false positive rate. This work has been tested through simulations and also using bench mark dataset for analysing the performance of the proposed algorithms. From the experiments conducted in this work using full features and selected features by applying the proposed classification algorithm, it is proved that the proposed framework detects the intruders more accurately and reduces the attacks leading to increase in packet delivery ratio and reduction in delay.

Keywords: intrusion detection system; IDSs; feature selection; classification; genetic algorithm based feature selection algorithm; GAFSA; neuro-genetic fuzzy classification algorithm; NGFCA; false positive rate; FPR; neuro-genetic framework.

DOI: 10.1504/IJRIS.2018.096202

International Journal of Reasoning-based Intelligent Systems, 2018 Vol.10 No.3/4, pp.224 - 232

Received: 19 May 2017
Accepted: 07 Aug 2017

Published online: 19 Nov 2018 *

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