I'm going to make a prediction that this will get massive hype and pomp surrounding it, and three or four (or whatever) years from now when the first big result sets roll in, it'll be discovered that it turns out it's not much more (perhaps not at all more), effective than random patrols in bad areas.

Why? Because we live in a dynamic, non-fate based world. Crime is a choice, and cannot actually be predicted (because destiny is not set). I can walk up to the door of a convenience store with a gun, stop, think, and turn back. Thus random patrols in known bad neighborhoods is guaranteed over a large dataset to be as effective as 'predictive' crime technology.