![IconGems-Print](images/IconGems-Print.png)
6 – 6 - Highlights in 'Bayesian Analysis': Stories to Tell
A Bayesian Hierarchical Model of Violent Criminal Threat
Francis O. Bunnin
The Alan Turing Institute
Violent criminals will often need to go through a sequence of preparatory steps before they can execute their plans. During this escalation process police have the opportunity to evaluate the threat posed by such people through what they know, observe and learn from intelligence reports about their activities. We customize a Bayesian hierarchical model to describe this process. This is able to propagate both routine and unexpected evidence in real time. The model structure comprises the latent threat state of an individual person of interest; the activities that person carries out relevant to an attack; and observable data produced by those activities. The model aims to support real-time decision making by security analysts. Specifically it aims to focus attention and allocate constrained resources on the cases that pose the greatest imminent danger.