Revolutionizing Network Security Stacked Generalization for Malicious Traffic Detection

View Sample PDF

Author(s): Harsh Jain (SRM Institute of Science and Technology, Ramapuram, India), Suhas Srinivas Lingam (SRM Universtiy, India), Azhagiri Mahendiran (SRM Institute of Science and Technology, Ramapuram, India)and Karthik Srinivasan (Saudi Electronic University, Saudi Arabia)
Copyright: 2026
Pages: 30
Source title: Recent Advances in Smart Communication Technologies for a Sustainable Future
Source Author(s)/Editor(s): Valentina E. Balas (Aurel Vlaicu University of Arad, Romania), Hari Mohan Pandey (Bournemouth University, UK), Mohd. Anshari Bin Ali (Universiti Brunei Darussalam, Brunei), Vineeta Singh (Chhatrapati Shahu Ji Maharaj University, Kanpur, India)and Alok Kumar (Chhatrapati Shahu Ji Maharaj University, Kanpur, India)
DOI: 10.4018/979-8-3373-3541-4.ch012

Keywords: Environment & Agriculture / Information Science Reference / Networking and Telecommunications / Sustainable Development

Purchase

View Revolutionizing Network Security Stacked Generalization for Malicious Traffic Detection on the publisher's website for pricing and purchasing information.

Abstract

The rising complexity of cyberattacks challenges traditional network security systems. To address this, we propose a layered generalization approach using multiple machine learning classifiers and XGBoost as a meta-learner to detect malicious network traffic more effectively. Stacking enhances prediction accuracy by combining base model strengths and reducing weaknesses. The classifiers are trained on real-world network traffic features, with XGBoost aggregating their outputs to improve detection. Its scalability, efficiency, and handling of imbalanced data suit real-world traffic's benign-malicious imbalance. Experiments on the CIC-IDS2017 dataset show that our stacked framework outperforms individual models in accuracy, precision, recall, and F1-score, significantly reducing false positives and negatives while strengthening network protection.

The IRMA Community

Research IRM

Revolutionizing Network Security Stacked Generalization for Malicious Traffic Detection

Purchase

Abstract

Related Content

IRMA Sponsors