Enhancing Intrusion Detection Based on Hybrid Dynamic Feature Selection Approach, Fine-Tuned GRU and Generative Adversarial Networks

DOI

10.2991/978-94-6463-805-9_5How to use a DOI?

Keywords

intrusion detection system; network security; dynamic feature selection; deep learning; machine learning; optimization algorithms; generative adversarial networks

Abstract

Intrusion detection systems (IDS) play an important role in protecting the networks from increasingly complex cyber-attacks, albeit their effectiveness is often compromised due to the high dimensionality of data and lack of optimization for gainfully trained models. To address these issues, we suggest a cutting-edge IDS framework that integrates a customized hybrid feature selection algorithm with an optimized Gated Recurrent Unit (GRU) model. The feature selection process employs a novel Hybrid EDA-BSO/SMA algorithm, combining Estimation of Distribution Algorithms (EDA), Brain Storm Optimization (BSO), and Slime Mould Algorithm (SMA) to effectively reduce data dimensionality by selecting the most pertinent features. Moth-Flame Optimisation (MFO) and Grasshopper Optimisation Algorithm (GOA) are then used to fine-tune the hyper-parameters of the GRU model. Furthermore, we use Generative Adversarial Networks (GANs) in the evaluation phase to gauge the IDS’s resilience and capacity for generalisation. Experimental evaluations on benchmark datasets show that our approach significantly enhances detection accuracy, precision, and recall, making it a robust solution for addressing real-world network security challenges.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Abid Dhiya Eddine
AU  - Ghazli Abdelkader
PY  - 2025
DA  - 2025/08/05
TI  - Enhancing Intrusion Detection Based on Hybrid Dynamic Feature Selection Approach, Fine-Tuned GRU and Generative Adversarial Networks
BT  - Proceedings of the First International Conference on Artificial Intelligence, Smart Technologies and Communications (AISTC 2025)
PB  - Atlantis Press
SP  - 32
EP  - 40
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-805-9_5
DO  - 10.2991/978-94-6463-805-9_5
ID  - DhiyaEddine2025
ER  -