Malware Classification from Memory Dumps Using Machine Learning, Transformers, and Large Language Models

This study investigates the performance of various classification models for a malware classification task using different feature sets and data configurations. Six models-Logistic Regression, K-Nearest Neighbors (KNN), Support Vector Machines (SVM), Decision Trees, Random Forest (RF), and Extreme Gradient Boosting (XGB)-were evaluated alongside two deep learning models, Recurrent Neural Networks (RNN) and Transformers, as well as the Gemini zero-shot and few-shot learning methods. Four feature sets were tested: All Features, Literature Review Features, the Top 45 Features from RF, and Down-Sampled with Top 45 Features. XGB achieved the highest accuracy of 87.42% using the Top 45 Features, outperforming all other models. RF followed closely with <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mathbf{8 7. 2 3 \%}$</tex> accuracy on the same feature set. In contrast, deep learning models underperformed, with RNN achieving 66.71% accuracy and Transformers reaching 71.59%. Down-sampling reduced performance across all models, with XGB dropping to 81.31%. Gemini's zero-shot and few-shot learning approaches showed the lowest performance, with accuracies of 40.65% and48.65 %, respectively. The results highlight the importance of feature selection in improving model performance while reducing computational complexity. Traditional models like XGB and RF demonstrated superior performance, while deep learning and few-shot methods struggled to match their accuracy. This study underscores the effectiveness of traditional machine learning models for structured datasets and provides a foundation for future research into hybrid approaches and larger datasets.