A survey on graph-based methods for malware detection

The widespread presence of malware is causing economic loss to organizations and individuals. For this reason, detecting malware has gained great interest as part of a computer security topic. Most current malware detection software uses signature-based methods to identify threats. However, this syntactic approach fails to identify variants of known malware or previously undiscovered malware. Since graphs represent a strong tool for modeling the sophisticated behaviors of malware and it is harder for an attacker to radically change the behavior of malware than to morph its syntactic structure, many graph-based methods are proposed to overcome the disadvantages of the traditional approach. In this paper, we present a survey of graph-based approaches for malware detection.