Analyzing Obfuscation Techniques for Evasion: A Case Study on Machine Learning-based Malware Detection

Un, Tugrul (2024) Analyzing Obfuscation Techniques for Evasion: A Case Study on Machine Learning-based Malware Detection. Masters thesis, Dublin, National College of Ireland.

Preview

PDF (Master of Science) Download (915kB) | Preview

Abstract

This research focuses on the intricate domain of malware obfuscation, a method utilized by malicious people to conceal the actual characteristics and operations of their code, thereby rendering its analysis and reverse engineering more difficult and time-consuming. Consequently, this enhances the malware’s capacity to elude detection and preventive methods. An area that is crucial but not thoroughly examined is the effect of these obfuscation techniques on malware detectors that rely on Machine Learning (ML). The primary objective of the research is to carry out a thorough examination of several obfuscation techniques, such as encryption, code obfuscation, and polymorphism, employed by attackers to conceal their malware. Having this comprehension is crucial for evaluating the present and possible future scenario of cyber risks. Additionally, the project aims to assess the impact of different obfuscation approaches on Static ML-based malware detectors. This study rigorously evaluates the impact of obscuring malware on the precision and efficiency of machine learning-based detection algorithms. An assessment of this nature is essential for uncovering the current capabilities and constraints of machine learning detectors when faced with advanced obfuscation techniques. Finally, the study aims to improve the identification and prevention methods in cybersecurity by identifying the weaknesses of machine learning-based malware detectors when faced with obfuscation attacks. The objective is to make a significant contribution to the area by suggesting ways that can strengthen the resistance of machine learning-based detectors against sophisticated and disguised malware threats. This would enhance cybersecurity defenses against constantly emerging malware issues.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Trinh, Anh Duong UNSPECIFIED
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 Artificial Intelligence
Depositing User:	Ciara O'Brien
Date Deposited:	30 May 2025 14:25
Last Modified:	30 May 2025 14:25
URI:	https://norma.ncirl.ie/id/eprint/7718

Actions (login required)

View Item