Malware Classification Using Machine Learning

Abstract

Recently, tools for generating malware have been spreading rapidly on the internet, making it easier for people without expertise to create malware. As a result, the number of malware variants is increasing quickly. To address this issue, it is crucial to classify malware quickly and accurately. However, malware variants are evolving to evade traditional malware-detecting methods based on signature pattern matching. To solve this problem, researches on detection of malware have been made in various fields. In the present study, we first propose a classification method to extract feature data from malware files that is applicable to machine learning, and then we classify malware through learning. Finally, we apply our classification method to sample data to evaluate performance and analyze the results.