Darknet Web URL Detection without URL Content Leakage
Jie Xu, Haoliang Lan · 2021 IEEE Conference on Telecommunications, Optics and Computer Science (TOCS) · 2021
Darknet is a kind of network that uses special software to communicate anonymously. The original purpose of anonymous communication is to protect users’ privacy. However, due to its untraceability, the darknet is used by criminals to engage in illegal activities. Detecting the URL of the darknet is the key step to control the darknet. According to the characteristics of the darknet web, this paper proposes a method to detect the dark web address by using URL entropy, an attribute independent of web content. URL entropy reflects the dispersion of URL characters and does not reveal the content of URL. Based on the URL entropy, this paper proposes a URL detection method for the darknet web URL, which does not disclose the URL content. Experiments show that the proposed URL detection model based on URL entropy has higher security and stability.