Fourier Transform as Feature Extraction for Malware Classification

Research efforts to develop malicious application detection algorithms have been a priority ever since the discovery of the first “viruses”. Fourier transform is used to extract features from binary files. These features are then reduced by random projection algorithm to create a set of low-dimensional features that are used to classify whether the application is malicious or not. A 99.6% accuracy was reached by Random Forest classifier, while processing various worms, trojan horses, viruses, and backdoors.