Organized crime

I built CRIMENET, an open-source network of 1,890 criminal organizations and 3,354 alliances and rivalries between them, extracted from 771 Wikipedia articles with an LLM pipeline.

Women are underrepresented in corruption networks in Brazil and Spain, but play similar roles to men except at the top. Gender doesn’t affect network resilience, and most differences are explained by their lower presence.

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

Interact with realistic simulations of corruption networks

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