A Novel Search Engine to Uncover Potential Victims for APT Investigations

Abstract

Advanced Persistent Threats (APT) are sophisticated and target-oriented cyber attacks which often leverage customized malware and bot control techniques to control the victims for remotely accessing valuable information. As the APT malware samples are specific and few, the signature-based or learning-based approaches are weak to detect them. In this paper, we take a more flexible strategy: developing a search engine for APT investigators to quickly uncover the potential victims based on the attributes of a known APT victim. We test our approach in a real APT case happened in a large enterprise network consisting of several thousands of computers which run a commercial antivirus system. In our best effort to prove, the search engine can uncover the other unknown 33 victims which are infected by the APT malware. Finally, the search engine is implemented on Hadoop platform. In the case of 440GB data, it can return the queries in 2 seconds.

These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.