Publications

Deep learning methods are used to predict properties of criminal networks.

We have demonstrated how structural properties of criminal networks and machine learning methods can be used to predict links and link features among actors engaged in nefarious activities.

Our results show that corruption networks share universal characteristics that are independent of social and cultural differences among countries.

Our research reveals that corruption runs in small groups that rarely comprise more than eight people, in networks that have hubs and a modular structure that encompasses more than one corruption scandal.