Malicious Domain Names Detection Algorithm Based on Statistical Features of URLs
Hong Ping Zhao, Zhiwen Chen, Rongjing Yan · 2022 IEEE 25th International Conference on Computer Supported Cooperative Work in Design (CSCWD) · 2022
Malicious domains often employ a number of illegal activities to threaten people's lives, including their privacy and property. As a result, the task of detecting malicious domain names has become a focus of attention. To solve the problem, we propose a novel malicious domain name detection algorithm which is statistically characterized by Uniform Resource Locator URL) characters. Firstly, statistical features of the given URL are extracted based on prior knowledge of domain name. Secondly, a decision tree is constructed based on URL characters to count various features. In addition, this decision tree is also used to detect the already generated domain names. Finally, the legal domain names and the domain names we generated are composed as a kind of dataset. The test results show that the proposed detection algorithm has the average accuracy rate of 90.31% and precision rate of 90.03%, which means a better performance on malicious domain name detection.