A new algorithm could help predict when the Islamic State (Isis) are planning their next attack.
By scanning the social media posts of Isis affiliated accounts, a team from the University of Miami say they can accurately guess where and when an attack will take place.
The method involves searching for "pro Isis language and statements" across a variety of languages and then building a map of recurring data. 100,000 accounts from around 200 groups were monitored on VK, a Russian social media site.
From this data, the team could build a "a statistical model aimed at identifying behavioral patterns among online supporters of ISIS and used this information to predict the onset of major violent events".
"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 team write. "The model suggests how the development and evolution of such aggregates can be blocked."
The study has been published in Science.
This article was originally published by WIRED UK
