This repository provides two web crawlers to label domain names using the McAfee API (https://www.trustedsource.org/sources/index.pl) and IP reputation using the TALOS API (https://talosintelligence.com/), respectively.

If you find this project useful for your research, please cite my paper.

@article{qin2024multi,
  title={Multi-Task DNS Security Analysis via High-Order Heterogeneous Graph Embedding},
  author={Qin, Meng},
  journal={arXiv preprint arXiv:2401.07410},
  year={2024}
}

• BeautifulSoup

Descriptions of the demonstration code are as follows.

1. To label the categories of a set of domains, put the domain list in 'data/domain_list.txt' and run 'demo_domain_label.py'. The program will label the (1) category (e.g., Malicious Sites- Parked Domain) as well as (2) risk level (e.g., High Risk) of each domain (using the McAfee API) and save the results in 'res/domain_labels.txt'. When the program continuously outputs ''-Retry-'', please stop the program and wait for a moment. After the waiting, you can start the program again, which can automatically skip the domains already labeled and continue to label the rest domains.

2. To label the reputation of a set of IP addresses, put the IP list in 'data/IP_list.txt' and run 'demo_IP_label.py'. The program will label the (1) email reputation as well as (2) web reputation (with 3 levels of Poor, Neutral, and Good) and save the results in 'res/IP_labels.txt'. When the program continuously outputs ''None'', please stop the program and wait for a moment. After the waiting, you can start the program again, which can automatically skip the IPs already labeled and continue to label the rest IPs.

3. An example domain name list (with 21,820 effective second-level domains) and an example IP list (with 67,751 IP addresses) are given in 'data/examples/example_domain_list.txt' and 'data/examples/example_IP_list.txt', repsectively. The corresponding labeled results are saved in 'res/examples/example_domain_labels.txt' and 'res/examples/example_IP_labels.txt', respectively.

If you have questions regarding this repository, you can contact the author via [mengqin_az@foxmail.com].