Static Detection of Malicious Code in Programs Using Semantic Techniques

In this paper, we propose a method for static detection of malicious intent of programs by expressing the properties of its source code through ontology. Our study concentrates on unraveling the traits of malware written in Java programming language by using semantic techniques. We have studied the source code of a known malware and identified the characteristics that make it a malicious one. We have expressed the said characteristics through an ontology that relates elements of programming language with each other. We have made the ontology expressive enough to detect the malware while providing a human-interpretable sketch. Code obfuscation enables the continued arms race between malware and malware detection systems. The work presented here also successfully found a solution to this problem.