The structure of criminal discussion forum networks on the dark net

As part of his MSc in Computer Science with Data Science specialization, Alain Fonhof analyzed real-world network data sets collected from online discussion forums on the dark net.

It turns out that network centrality measures can succesfully identify and characterize key players on the forum, distinguishing automatically between regular users, managers and technical moderators. The macro level properties and distributions of the network furthermore revealed forum-specific regulations and practices, such as anti-lurker and anti-law enforcement policies as well new user application guidelines. In addition, evidence was found suggesting that active users form an elite that participates in more specialized discussions on the forum.

The findings in this paper pave the way for automatically understanding online forums, without the need of understanding its textual contents. This is important in for example law inforcement. Indeed, Alain did his master thesis graduation project in collaboration with the Dutch Police. He presented his findings at the Complex Networks conference in Cambridge in December 2018. The work is summarized in the following paper: