Metamorphic malware categorization using co-evolutionary algorithm

Malware is a malicious code which intends to harm computers and networks. As malware attacks become pervasive, the security policy of computers is more critical and it is so important to have a well-defined process to detect malware. However to avoid detection of the malware, various concealment strategies are invented regularly. Metamorphism is a strategy in which malware change their codes on each infection, meanwhile keeping the functionality unchanged. We focus on these types of malware due to their complex behaviors. In this work we concentrate on Visual Basic Script (VBS) malware and propose a detection mechanism for metamorphic malware. Regarding the great ability of evolutionary algorithms, here, we employ a Co-evolutionary-based architecture to tackle the graph isomorphism problem to be able to detection metamorphic malware based on their semantic graph. The experimental results confirm the efficiency of the proposed method regarding other state of the art ones in the literature.