advertisement
advertisement

Wondering What Will Terrorists Do Next? There’s an App for That

The Israel-Lebanon border is normally one of the world’s nastiest conflict hotspots which is a headache for security forces and the regional governments. Which is where an odd app can help. It predicts how bad (or good) things may get.

Israel lebanon

advertisement

The Israel-Lebanon border is normally one of the world’s nastiest conflict hotspots which is a headache for security forces and the regional governments. Which is where an odd app can help. It predicts how bad (or good) things may get.

SOMA, the Stochastic Opponent Modeling Agents, is the app (or rather suite of apps) we’re talking about. It was developed by the University of Maryland’s Laboratory for Computational Cultural Dynamics, and it’s accurate enough to have predicted that the Israel-Lebanon border (a segment of which is shown above) would be relatively calm for terrorist activity right now. Which it is.

How the heck can some software do that? The answer is thanks to some very cleverly-written algorithms, and vast historic data sets. Think of SOMA as a chips-and-bits equivalent to Tom Clancy’s CIA analyst character Jack Ryan: Analytically pouring over data on how terrorist groups have behaved they way they did before, factoring in known relationships between persons, environmental situations and current events, and working out how the groups may behave in the future. The software code essentially looks at previous behavior and what appeared to motivate it, and assigns a probablility to how what particular terrorists might do next. It does so by weighing up the meanings and implications of thousands of piece of data–known facts, intelligence information, and weaving them all together probabilistically to come to an answer (that’s the “stochastic” part.) It can perform this trick with far more pieces of data and variables than a human analyst could, at least in a reasonable time period.

The datasets that feed SOMA include the Minorities at Risk Organizational Behavior database, also at the University of Maryland. In fact, MAROB helped SOMA predict the current quiet stat on the Lebanese border.

Why is this interesting? Because though SOMA obviously can’t be relied on to make firm decisions, since a probability-based system and the parties in question may simply act other than as predicted, it will be incredibly useful to senior officers in the security forces of hotspots like the Israel/Lebanon border. And it has other uses too, like predicting the impact of new policies designed to curb the growth of poppy for heroin in Afghanistan.

advertisement

Apps, it seems, are wending their way deeper and deeper into global life every day.

[Via ForeignPolicy]

About the author

I'm covering the science/tech/generally-exciting-and-innovative beat for Fast Company. Follow me on Twitter, or Google+ and you'll hear tons of interesting stuff, I promise.

More