Evaluating Next-Generation Firewalls Using Machine Learning-Based Hybrid Feature Selection for Threat Detection and Risk Mitigation

Trishna Panse, Kailash Chandra Bandhu

2025

Abstract

Securing modern networks against evolving cyber threats requires robust and intelligent systems. This research introduces a novel machine learning-based framework designed to enhance Next-Generation Firewalls (NGFWs) by improving their ability to manage traffic, detect intrusions, and mitigate risks. We propose a Random Forest (RF)-based classification approach, leveraging a hybrid feature selection strategy that combines filter-based ranking (using Information Gain Ratio) with wrapper-based validation (k-means clustering). In our proposed research we use CSE-CIC-IDS2018 dataset (Canadian Institute for Cybersecurity - Intrusion Detection System 2018) - is one of the most widely used benchmark datasets for network intrusion detection system (NIDS) research., KDD'99 dataset and assess the effectiveness of a Next-Generation Firewall (NGFW) in replicating and improving advanced threat-handling mechanisms. Our findings suggest that integration of intelligent machine learning models to detect the threats in NGFWs and achieve more than 98% of the accuracy in threat detection. It recognizes the powerful combined effect of the machine learning based Intrusion Detection System (IDS) functionalities in NGFWs, which provides a scalable and very effective solution for dynamic network security.

Paper Citation

in Harvard Style

Panse T. and Bandhu K. (2025). Evaluating Next-Generation Firewalls Using Machine Learning-Based Hybrid Feature Selection for Threat Detection and Risk Mitigation. In Proceedings of the 1st International Conference on Research and Innovations in Information and Engineering Technology - Volume 1: RITECH; ISBN 978-989-758-784-9, SciTePress, pages 149-156. DOI: 10.5220/0014275000004928

in Bibtex Style

@conference{ritech25,
author={Trishna Panse and Kailash Chandra Bandhu},
title={Evaluating Next-Generation Firewalls Using Machine Learning-Based Hybrid Feature Selection for Threat Detection and Risk Mitigation},
booktitle={Proceedings of the 1st International Conference on Research and Innovations in Information and Engineering Technology - Volume 1: RITECH},
year={2025},
pages={149-156},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0014275000004928},
isbn={978-989-758-784-9},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Innovations in Information and Engineering Technology - Volume 1: RITECH
TI - Evaluating Next-Generation Firewalls Using Machine Learning-Based Hybrid Feature Selection for Threat Detection and Risk Mitigation
SN - 978-989-758-784-9
AU - Panse T.
AU - Bandhu K.
PY - 2025
SP - 149
EP - 156
DO - 10.5220/0014275000004928
PB - SciTePress