Malware Detection via Classifying with Compression

摘要: The proliferation of malware has been causing great harm to computer and information systems. Traditional signature-based approaches fail detect obfuscated unknown malware. We present a preliminary study on classifying with compression program instructions for detection. code structure was utilized compress. disassembled is converted its intermediate representation. extract the opcode form stream prediction by partial matching (PPM). binaries are classified this statistical algorithm. experiment shows that our method can efficiently high accuracy low false positive rate.