ML-Based Zero-Day Attack Detection

Mathew, Ans Maria (2024) ML-Based Zero-Day Attack Detection. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (684kB) | Preview
Preview	PDF (Configuration Manual) Download (264kB) | Preview

Abstract

In this paper discussing a new machine learning based approach and architecture for intrusion detection system (IDS) for detecting zero-day attack. By employing IoT23 dataset, the research employs supervised learning with Random Forest; unsupervised learning through Isolation Forest; deep Neural Network (DNN) for the high criticality data. Expanding the dataset by Smote as well as feature scaling helped in achieving a good performance by the models. Considering the achieved outcomes, having high values of accuracy level, as well as pioneers’ high level of precision and recall, it is worth to concentrate on DNN as the most effective and accurate variant with over 99% of accuracy. Validity issues such as privacy and fairness were considered in the study. This multilayering makes it quite effective to hold up as a model for practical applications in the field of cybersecurity.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Salahuddin, Jawad UNSPECIFIED
Uncontrolled Keywords:	Zero-day attack; Intrusion Detection; Machine Learning; Anomaly Detection; Supervised learning; Unsupervised learning; Neural Network
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science Q Science > QA Mathematics > Computer software > Computer Security T Technology > T Technology (General) > Information Technology > Computer software > Computer Security Q Science > Q Science (General) > Self-organizing systems. Conscious automata > Machine learning
Divisions:	School of Computing > Master of Science in Cyber Security
Depositing User:	Ciara O'Brien
Date Deposited:	23 Jul 2025 15:07
Last Modified:	23 Jul 2025 15:07
URI:	https://norma.ncirl.ie/id/eprint/8225

Actions (login required)

View Item