arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2212.03641cs.IRcs.CR
收藏

THREAT/crawl:一种可训练、高度可重用和可扩展的自动化方法和工具,用于抓取犯罪地下论坛

THREAT/crawl: a Trainable, Highly-Reusable, and Extensible Automated Method and Tool to Crawl Criminal Underground Forums

Michele Campobasso, Luca Allodi

摘要:

收集关于地下犯罪社区的数据对于安全研究和安全行动都非常有价值。不幸的是,这些社区生活在一个难以渗透的各种在线论坛的星座中,可能会采取爬行监控对策,并需要为每个不同的社区开发专门的抓取器,这使得这项工作在技术上越来越具有挑战性,并且可能昂贵。为了解决这个问题,我们提出了THREAT/crawl,一个高度可重用的爬虫程序,可以学习广泛的(任意)论坛结构的方法和原型工具,可以保持在雷达下的爬行活动,并可以扩展和配置在用户的意愿。我们展示了威胁/爬行能力,并对我们的原型进行了一系列活跃的、实时的地下社区的初步评估。

英文摘要:

Collecting data on underground criminal communities is highly valuable both for security research and security operations. Unfortunately these communities live within a constellation of diverse online forums that are difficult to infiltrate, may adopt crawling monitoring countermeasures, and require the development of ad-hoc scrapers for each different community, making the endeavour increasingly technically challenging, and potentially expensive. To address this problem we propose THREAT/crawl, a method and prototype tool for a highly reusable crawler that can learn a wide range of (arbitrary) forum structures, can remain under-the-radar during the crawling activity and can be extended and configured at the user will. We showcase THREAT/crawl capabilities and provide prime evaluation of our prototype against a range of active, live, underground communities.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑