Malicious HTTP communication detection based on access graph analysis

Abstract

HTTP is recognized as the most widely used protocol on the Internet when development of applications is transferred more and more onto the web. Therefore, malicious developers trend to exploit HTTP as a communication media environment to spread forbidden actions. Detection of malicious HTTP communication is a really huge challenging job since the malicious HTTP communication is transparently merged with other type of HTTP traffic. Hence, in this paper, based on previous studies and by analyzing HTTP communication behavior through access graph, a new method is proposed to detect malicious HTTP communication. Experiment results are promising since detection precision constitutes a proportion of 86.96%.