A New Program Classification Method Based on Binary Instrumentation and Instruction Flow Feature Extraction

With the development of Internet technology, more and more unknown programs appears in the network environment, the detection and classification technology is increasingly becoming an important technology in the field of information security. This paper presents a new method of program classification using binary instrumentation, dynamic instruction flow feature extraction, auto feature selection and Naive Bayes classifier technology. Finally we use two-class classification, five categories, and twenty-class classification experiments to prove the correctness of the method and present the future direction for improvement of the method.