Dynamic Innate Immune System Model for Malware Detection

Abstract

Malware stand for Malicious Software became a major threat facing the massive amount of data transmitted through the internet and the systems holding that data. Malware detection is the process of identifying the malicious behavior or object as malware. Many methods used to do the detection process, these methods are varied depending on the process used by the detector -anti virus or anti malware is a commercial name of detectors. Signature base, behavior base and specification base. Increasing the detection accuracy is the main goal of researchers in the last decade. In this paper we introduce a dynamic malware detection model by applying the innate immune system to improve the detection accuracy. The proposed model applied to the portable executable file representation by extracting the API call logs from new installed windows environment due to the wide spread of this type of files in different platforms. The results of the experiments show a better detection accuracy of the proposed model for known malware and promising improvement on the new unknown malware and polymorphic malware.