Design Secure Authentication Methods to Prevent Unauthorized Access to Critical Systems and Networks

Tharun D., V. Madhu Sudhan, C. Hemalatha, Abirami G.

2025

Abstract

As cyber threats become increasingly more complex, securing the communication of network communication is a major challenge. One of the reasons for that is that traditional security measures are not able to adjust with constantly changing attack tactics, thus, it is necessary to incorporate ML driven approaches to improve network security. In this research study, we examine the efficiency of different ML model for handling malicious activities in network traffic employing UNSW-NB15 dataset. NS-LR, NS-SVM, NS-KNN, NS-MLP, NS-GBM and a Stacking Classifier is applied for comparison with each other. A complete feature correlation analysis and multiple class classification is conducted to evaluate the model performances with a aid of metrics as accuracy and AUC-ROC. The conducted results show Stacking Classifier’s superiority over other models with near perfect accuracy that determines it as a good choice for real time network intrusion detection systems. In contrast, MLPs limited the classification performance and led to additional training needed. This work demonstrates the feasibility of application of ML driven schemes for the protection of contemporary networks against highly sophisticated cyber attacks. The inclusion of adversarial learning techniques and the integration with explainable AI (XAI) towards real time deployment will then be taken as future work.

Paper Citation

in Harvard Style

D. T., Sudhan V., Hemalatha C. and G. A. (2025). Design Secure Authentication Methods to Prevent Unauthorized Access to Critical Systems and Networks. In Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25; ISBN 978-989-758-777-1, SciTePress, pages 95-100. DOI: 10.5220/0013908700004919

in Bibtex Style

@conference{icrdicct`2525,
author={Tharun D. and V. Sudhan and C. Hemalatha and Abirami G.},
title={Design Secure Authentication Methods to Prevent Unauthorized Access to Critical Systems and Networks},
booktitle={Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25},
year={2025},
pages={95-100},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013908700004919},
isbn={978-989-758-777-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 1st International Conference on Research and Development in Information, Communication, and Computing Technologies - ICRDICCT`25
TI - Design Secure Authentication Methods to Prevent Unauthorized Access to Critical Systems and Networks
SN - 978-989-758-777-1
AU - D. T.
AU - Sudhan V.
AU - Hemalatha C.
AU - G. A.
PY - 2025
SP - 95
EP - 100
DO - 10.5220/0013908700004919
PB - SciTePress