How Scientists Are Using Mathematical Models to Predict ISIS Attacks
They’re fighting terrorism with an alogrithm.
Acts by so-called “lone wolf” terrorists who adhere to ISIS’s medieval beliefs have prompted researchers to seek new ways to forecast the next Orlando or San Bernardino. In a recent study published in Science, a group of scientists detailed how they created mathematical models that may one day lead to an early warning system for future attacks.
In their study summary the authors explained support for groups like ISIS or al-Qaeda survives “globally online despite considerable external pressure.” That support, they said, “may ultimately inspire acts by individuals having no history of extremism, membership in a terrorist faction, or direct links to leadership.”
Researchers examined an eight month period of online activities in forums supporting terrorist ideologies. They found an evolving “ecology” in which the groups self-organized and grew little by little on a daily basis.
Using this data, they put together a mathematically-supported theory to describe how “self-organized” small groups assembled on Facebook or on Russia’s Vkontakte (VK) before significant terrorist actions then adapted—even after a social network shut them down—to stay together.
A New York Times report about the study detailed skeptical responses from experts on terrorism like J.M. Berger, author of ISIS: The State of Terror. Berger told the Times the team had come up with “an interesting approach.” He also admitted it was “potentially valuable” but questioned its usefulness, stating that “to jump ahead to the utility of it, I think, takes more work.”
The lead study author, Dr. Neil Johnson, used examples from nature to explain some of his team’s conclusions to the Times. The way terrorist groups switch from talk to action, Johnson told the paper, is a lot like the spontaneous group movements of a flock of birds or a school of fish. “There’s no one fish saying, ‘Hey, I want everyone to be about five inches away from someone else, and we’re going to have this shape,’” Johnson said.
Johnson and his team found that prior to the Islamic State’s sudden siege of Kobani in September, 2014, there was a notable rise in newly-created pro-ISIS groups on some social networks.
The team applied their predictive formulas to that known sequence of events and found that if they’d been able to use their data in the past, they could’ve predicted Kobani—a positive step toward one day possibly using physics and math to determine where the next terrorist strike may occur.
While it’s wise to be skeptical of a study based on a relatively limited batch of data, acts by self-radicalized lunatics like Omar Mateen make it clear that counter-terrorist agencies worldwide are going to need every tool available to keep the malevolent spread of groups like the Islamic State at bay.
h/t New York Times