Intelligent malware detection based on file relation graphs

摘要: Due to its damage Internet security, malware and detection has caught the attention of both anti-malware industry researchers for decades. Many research efforts have been conducted on developing intelligent systems. In these systems, resting analysis file contents extracted from samples, like Application Programming Interface (API) calls, instruction sequences, binary strings, data mining methods such as Naive Bayes Support Vector Machines used detection. However, driven by economic benefits, diversity sophistication significantly increased in recent years. Therefore, calls much more novel which are capable protect users against new threats, difficult evade. this paper, other than based we study how relation graphs can be propose a Belief Propagation algorithm constructed detect newly unknown malware. A comprehensive experimental real large collection Comodo Cloud Security Center is performed compare various approaches. Promising results demonstrate that accuracy efficiency our proposed method outperform alternate techniques.