Sequence Aware Machine Learning for Attack Pattern Detection in SIEM Systems

Saboo, Devanshu Bharat (2025) Sequence Aware Machine Learning for Attack Pattern Detection in SIEM Systems. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (1MB) | Preview
Preview	PDF (Configuration Manual) Download (644kB) | Preview

Abstract

Traditional Security Information and Event Management (SIEM) systems often detect threats by evaluating isolated log events, overlooking the sequential relationships that define complex cyberattacks. This research addresses that limitation by developing a sequence based anomaly detection tool that can identify dependencies or patterns of attacks between multiple single event IDs within system logs. The goal is to demonstrate that analyzing event patterns can more accurately identify potential security incidents than event level analysis alone.

The project employs classical sequence mining techniques to discover frequent and anomalous event chains in log data collected through a custom Python based SIEM application. This approach emphasizes efficiency while avoiding the heavy computational cost of deep learning methods. Preliminary evaluation aims to reveal whether certain event sequences are consistently followed up by abnormal system activity, thereby improving the quality and prioritization of alerts.

The expected contribution is a lightweight, explainable and sequence aware detection method that enhances SOC analysts ability to recognize coordinated attack behaviours. This work focuses on pattern extraction and proof of concept validation by using sequence mining with machine learning for adaptive real time threat detection.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Jayasekera, Evgeniia UNSPECIFIED
Uncontrolled Keywords:	SIEM; log analysis; anomaly detection; sequence mining; cybersecurity
Subjects:	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:	04 Sep 2026 09:37
Last Modified:	04 Sep 2026 09:37
URI:	https://norma.ncirl.ie/id/eprint/9823

Actions (login required)

View Item