Logistic regression for polymorphic malware detection using ANOVA F-test

摘要

Today's world is rapidly moving towards digitization. In this context, protecting and safeguarding the digital resources is very crucial for a large organization or a country. Digital resources are attacked and virtually brought down using malware. One of the strategies to defend against malware is searching for a pattern inside them. These patterns become the signature for a malware and they are deployed into a security system for detection. But the traditional signature generation techniques fail against polymorphic malware, which change their form after every infection. In this paper, we propose a defense system which uses, Logistic regression with Anova F-Test and snort IDS to thwart these polymorphic malware. Logistic regression with Anova F-Test has achieved 97.7% accuracy.