© 2024 WLRN
MIAMI | SOUTH FLORIDA
Play Live Radio
Next Up:
0:00
0:00
0:00 0:00
Available On Air Stations

UM Physicist Leads Charge For Algorithm To Find Terrorists On The Web

New online ecology of adversarial aggregates: ISIS and beyond
Horizontal bars illustrate timelines of some typical pro-ISIS aggregates or web groups, according to the data gathered by the UM research team.

 

The ability to predict when talks of terrorism on social media will manifest into attacks is one step closer to reality.

 

A University of Miami team of physicists published a study in the journal Science describing a mathematical algorithm that takes a new approach to monitor ISIS conversations online and can help predict possible attacks.

 

According to their findings, "sudden escalation in the number of ISIS-supporting ad hoc web groups ("aggregates") preceded the onset of violence in a way that would not have been detected by looking at social media references to ISIS alone."

 

The novelty of this new method is that it is not about looking at the total volume of Google searches but rather it focus on the proliferation of smaller, ad hoc groups linked through social media sites and the increases of activity by its members.

 

"Facebook shuts them down very quickly but the equivalents of Facebook around the world, they are just slower to shut them down," Johnson explained. "Each one of those aggregates, self-police from within."

 

Johnson, who was the leader of the study, said that people prefer to congregate in many small groups, as opposed to a few large ones.

 

“Animals, fish, birds, etc. they move into groups and they break up when there’s no kind of top-down organization. That’s how collections of living objects do things, and I don’t see any good argument for thinking that humans would be any different," Johnson said. "And that’s what we find - they’re not different.”    

 

Johnson said the study cannot anticipate when a specific terror event would occur, but the findings can be used to get a better sense of periods where they are more likely.

 

“They can’t tell you on Tuesday you’ll have a heart attack. But they tell you that you are now in a high-risk group. So it’s very much that level of prediction,” he said.

 

 

 

More On This Topic