Enhancing Malware Detection Accuracy through Graph Based Model
K. Muthumanickam *
Research Scholar, Department of Computer Science and Engineering, Pondicherry Engineering College, Puducherry – 605 014, India.
E. Ilavarasan
Department of Computer Science and Engineering, Pondicherry Engineering College, Puducherry – 605 014, India.
*Author to whom correspondence should be addressed.
Abstract
Malicious malware is a serious threat to end-user in the Internet. Run-time analysis of a program execution behavior is widely used to classify malware’s activities especially when its signature is not obtainable. Towards this end, most of the existing run-time malware detection techniques make use of the information available in the Application Programming Interface call sequence in Windows platform. This paper suggests a novel malware revealing model based on graph model by capturing system calls during the execution of a suspected executable. The implementation results confirm that the proposed call graph model has better detection accuracy rate and also solves the scalability problem when it is compared to existing methods.
Keywords: Graph model, isomorphism, malware, virtualization