作者: Abedelaziz Mohaisen , Omar Alrawi
关键词:
摘要: Malware family classification is an age old problem that many Anti-Virus (AV) companies have tackled. There are two common techniques used for classification, signature based and behavior based. Signature uses a sequence of bytes appears in the binary code to identify detect malware. Behavior artifacts created by malware during execution identification. In this paper we report on unique dataset obtained from our operations classified using several machine learning behavior-based approach. Our main class interested classifying popular Zeus For its 65 features robust identifying families. We show like file system, registry, network can be distinct families with high accuracy - some cases as 95 percent.