Title: A combined feature selector using Jaya and differential evolution to improve the classification accuracy for dataset of intrusion detection system

Authors: S. Appavu Alias Balamurugan; A.S. Karthik Kannan; Millie Pant

Addresses: Department of Applied Mathematics and Scientific Computing, Indian Institute of Technology Roorkee (IITR), Saharanpur Campus, India ' Department of Applied Mathematics and Scientific Computing, Indian Institute of Technology Roorkee (IITR), Saharanpur Campus, India ' Department of Applied Mathematics and Scientific Computing, Indian Institute of Technology Roorkee (IITR), Saharanpur Campus, India

Abstract: Cyberattacks are considered one of the largest threats to data security in this digital age. In the overall strategy for thwarting cyberattacks, intrusion detection systems (IDS) play a very important role. A high dimensional data flow poses a significant challenge for IDS when investigating all aspects. IDS's success rate is reduced as a result of the increases in computation cost. Feature selection in intrusion detection is proposed as a combined self adapted Jaya optimisation algorithm. The goal of the proposed work is to maximise the classification accuracy (success rate) and to minimise the feature selection ratio by maximising the fitness function. Three benchmark datasets (UNSW-NB15, KDDCUP '99 and NLS KDD) were used to verify the proposed method performance. With reference to the analysis of comparison made, the proposed method outperforms than the existing methods.

Keywords: combined feature-selector; classifier; system for intrusion detection; Jaya optimisation; differential evolution.

DOI: 10.1504/IJISTA.2024.143252

International Journal of Intelligent Systems Technologies and Applications, 2024 Vol.22 No.4, pp.405 - 426

Accepted: 24 Jul 2024
Published online: 11 Dec 2024 *

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