Physicists associated with the University of Miami created an algorithm to track patterns on social media in an attempt to predict future ISIS attacks. With the Islamic State’s reliance on social media, Dr. Johnson led Miami colleagues in searching for pro-Islamic State posts in multiple languages on the social site Vkontakte each day from mid-2014 through August 2015. Mentions of beheadings, blood baths, and posts that were violent in nature were observed.
Mentions of beheadings, blood baths, and posts that were violent in nature were targeted. The study was published Thursday in the journal Science.
Vkontakte is a Russian-oriented social media service and Europe’s most widely used social platform. Investigators included experts who spoke an array of languages such as Arabic, Chinese, French, German, Russian, and Spanish because Vkontakte, unlike Facebook, doesn’t police pro-Islamic posts. The social media platform has a diverse 350 million users, many of whom are Chechen radicals who have been targets of Islamic State propaganda.
The social media platform has a diverse 350 million users, many of whom are Chechen radicals who have been targets of Islamic State propaganda.
The study concentrated its efforts on small amorphous Islamic State groups and found nearly 200 of them with 100,000 members combined. These small groups indicate surges of activity and if traced, could lead authorities to suspected targets. Although this tracking data can’t prevent an individual attack—Orlando or San Bernardino for example—it could clue government agencies into ripe conditions for one.
Dr. Johnson said tracking terrorists on social media is much like, “the way transitions happen is like a flock of birds, a school of fish,” and “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.”
Professionals who study online terrorism and online communication said that the novel research was instructive and praised the scientists’ efforts. But even with this data, executions of the attacks are difficult to anticipate and more study was needed to rely on such a model, experts advised.
“This is an interesting approach, this is a potentially valuable approach, and more research should be done on the approach,” said J. M. Berger, research fellow at George Washington University’s Program on Extremism and the co-author of “ISIS: The State of Terror.” “But to jump ahead to the utility of it, I think, takes more work.
The full report was published in The New York Times.