Publication

Abstract : Detecting malicious software, known as malware, is crucial in cybersecurity due to the constantly evolving threats and the ways in which malware tries to avoid detection. This research investigates the efficacy of deep learning models for multi-class malware classification using the MaleVis image dataset. Research on using transformer architectures for malware detection and classification is limited. The proposed study explores using the Convolutional Vision Transformer (CvT) for detecting malware, comparing its performance with the Vision Transformer (ViT) and the pre-trained Convolutional Neural Network, EfficientNet B0. Each model is fine-tuned on the MaleVis dataset to distinguish between different malware categories and benign samples. A comprehensive assessment using multiple evaluation metrics suggests CvT outperformed the other models, with an F1-Score of 0.96054. ViT followed closely with a score of 0.95821, while EfficientNet B0 scored 0.87386. The research aims to contribute to cybersecurity advancements by leveraging modern deep learning techniques for enhanced malware detection.