Any such system is/would be potentially very dangerous. Crime data is not the same thing as crime. Populations that are over-policed are be disproportionately represented in any such data set, leading to higher prediction of crime, leading in turn more over-policing (feedback loop). I implore anyone attempting to build such a system to consider the serious issue of machine bias and it's implications in the real world.

See this tutorial given at this years NIPS machine learning conference: http://mrtz.org/nips17/#/

Potential dangers of such a film are highlighted in the film Minority Report. https://en.wikipedia.org/wiki/Minority_Report_(film)

I also wrote a little blog post about that very thing back in 2011. In addition to using a Markov chain approach, also took a look at it from an information entropy perspective. And the code is also in R, to boot! http://bayesianbiologist.com/2011/12/31/uncertainty-in-marko...

Indeed. It's not a transcript of the talk, but here are the accompanying materials that we worked with during the class. http://bayesianbiologist.com/2012/10/19/introduction-to-baye...

WikiRaces! https://twitter.com/wikiraces

Yay for steam locomotive! I think of it as a handy reminder from your sys-admin "don't drink and ssh!"

That same sys-admin probably won the Crash And Compile drinking game at DEF CON. Don't listen to him.