Harnessing Deep Features and Machine Learning for Malware Image Classification

Thokala, Poojitha (2024) Harnessing Deep Features and Machine Learning for Malware Image Classification. Masters thesis, Dublin, National College of Ireland.

Preview	PDF (Master of Science) Download (894kB) | Preview
Preview	PDF (Configuration Manual) Download (1MB) | Preview

Abstract

The increase of sophisticated malware poses a critical threat to individuals, organizations, and critical infrastructures, which highlights the urgent need for better and robust detection mechanisms. Traditional malware detection methods such as signature-based and heuristic methods, struggle to handle particularly in case of constantly evolving nature of malware, obfuscation, polymorphism techniques and imbalanced datasets. This often results in poor performance leaving the systems vulnerable.

This method first transforms malware binaries into grayscale images, allowing CNN to extract key spatial features. Support Vector Machines (SVMs) and Random Forests (RFs) are used to classify the features which are further combined through a learning strategy to improve accuracy and robustness. This study shows the advantages of combining deep learning for automated feature extraction with traditional machine learning for precise classification.

The results show that this hybrid method could be a practical and scalable solution for modern malware detection resulting in better accuracy and efficiency. This methodology was validated on the Malimg dataset, which has 25 malware families, achieving accuracy of 97.2%. Strong predictive performance is shown through the confusion matrices, but minor trends of misclassification, that require further refinement are also shown.

Item Type:	Thesis (Masters)
Supervisors:	Name Email Sahni, Vikas UNSPECIFIED
Subjects:	Q Science > QA Mathematics > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Electronic computers. Computer science T Technology > T Technology (General) > Information Technology > Cloud computing 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 Cloud Computing
Depositing User:	Ciara O'Brien
Date Deposited:	17 Jul 2025 12:40
Last Modified:	17 Jul 2025 12:40
URI:	https://norma.ncirl.ie/id/eprint/8160

Actions (login required)

View Item