Although political scientists and economists have been working for decades on the question of whether and under what circumstances political conflicts will turn into crises, forecasting armed conflicts is notoriously difficult. For around ten years, however, researchers around the world have been working on using data analysis and machine learning to make this prediction.
However, crisis prediction becomes particularly difficult in cases in which violence breaks out for the first time – without armed conflicts having previously occurred. Developments such as the Arab Spring of 2011, but also Russia’s sudden attack on Ukraine, are usually not good for models that work with conflict history and structural data.
Hannes Müller from the Barcelona School of Economics and Christopher Rauh from the University of Cambridge have therefore joined forces in the conflictforecast.org project to specifically tackle these “difficult cases” of conflict prediction. Their idea: They let algorithms evaluate newspaper articles – and then use this evaluation to calculate a conflict probability. This enables them, among other things, to update the forecast on a monthly basis.
Editors of the innovation magazine Technology Review discuss important facts and significant absurdities, small anecdotes and big connections.
In the new podcast episode of MIT Technology Review, TR editor Wolfgang Stieler talks to the researchers about how this works (nerd spoiler: this podcast contains a brief introduction to random forests), what the limitations of the method are, and what comes of them predictions follow.
“We’re practically looking for a needle in a haystack,” says Hannes Müller. “But even if we are wrong in most cases where we warn of an impending civil war, it is still worth intervening in all such cases. You have to realize that there are currently 80 million people on the run are facing wars. 30 million of them are children. A civil war is something like a nuclear accident on a political level.”
The whole episode as an audio stream (RSS feed) to listen to and download:
You can read more about risk analysis using artificial intelligence in the next issue of the MIT Technology Review. The new issue is out on September 28th. can be ordered in the heise shop and from 29.9. available in well-stocked newsagents.
=== Ad / Sponsor Notice ===
This podcast is sponsored by msg. At msg, as a specialist (m/f/d) in software engineering or IT consulting, you will drive the digital transformation of entire industries. Discover exciting IT jobs, the best individual development opportunities and flexible working hours with overtime compensation for your healthy work-life balance! More information at: karriere.msg.group
=== Ad / Sponsor Notice ===
To home page