Mining International Political Norms from the GDELT Database

Researchers have long been interested in the role that norms can play governing agent actions multi-agent systems. Much work has done on formalising normative concepts from human society and adapting them for government of open software systems, simulation processes artificial societies. However, there comparatively little applying MAS mechanisms to understanding society. This investigates this issue context international politics. Using GDELT dataset, containing machine-encoded records events extracted news reports, we bilateral sequences inter-country applied a Bayesian norm mining mechanism identify best explained observed behaviour. A statistical evaluation showed model fitted data significantly better than probabilistic discrete event model.