Title: Improving Binary Malware Detection Accuracy with Feature-Engineered Synthetic Data and Reduced False Negatives using MATLAB

Date Published: 7-12-2024

Authors: C. Romero

Keywords: Binary Classification, Feature Engineering, Malware Detection, Machine Learning, Synthetic Data

Abstract: Traditional malware detection methods struggle to keep pace with the evolving threat landscape. This research proposes a novel approach to improve binary malware classification accuracy using feature-engineered synthetic data generated in MATLAB. The proponent’s method focuses on extracting informative features from real malware samples to create synthetic data that closely resembles real-world malware behavior. This data is then used to train a machine learning model for classifying malware and clean files. The results demonstrate that the proposed method significantly improves binary malware detection accuracy, particularly by reducing false negatives or missed malwares. Additionally, the study investigates the model's ability to generalize to unknown malware threats. By applying synthetic data generation and feature engineering, this research offers valuable insights for developing more robust and reliable malware detection systems capable of handling the ever-changing threat landscape.