GDELT

To begin, the new school points out that political conflict is not best characterized as something that countries catch (or experience): the implicit epidemiological analogy is passé. Instead the new generation of scholarship recognizes that governments often face multiple challengers fighting for the same cause and/or very different causes, and that these challenges vary across both space and time. As such, these theorists have been disaggregating the study of civil conflict across actors, tactics, space and time.

Nice

I bet the author is not crazy about “tipping points” either.

Anyway. Here is also a blog post by one of the authors of the article (W. Moore). GDELT? It is a dataset containing 200 million events in the world which can be  mined / analyzed to predict conflicts, in the style popularized by Moore. A similar dataset ICEWS is apparently used by the US gov now, and a key contributer to that project happens to be a PhD student of Moore.