The largest and deadliest terrorist attack on US soil since 9/11 has shaken the nation, and many are wondering what the government is doing to prevent them. True enough, terrorist attacks all over the world have been surprising everyone, which could mean terrorist intelligence is far more advanced. Luckily, the FBI and several intelligence agencies have reportedly come up with an algorithm to help predict future attacks.
According to Engadget, a physicist from the University of Miami has come out with an algorithm he and his team have used to look over questionable social media posts. The team of Dr. Neil Johnson observed a year's worth of posts on Vkontakte, a Russian-based social network site. The data they have taken has allowed them to build "a statistical model aimed at identifying behavioral patterns among online supporters of ISIS."
Rather than quickly going through all posts, the model looks at how smaller, self-organized groups suddenly appear online before real-world campaigns. Following these groups could potentially make it easier for intelligence agencies to make a stop at terrorist attacks.
As Mashable explains, it is like looking at "the smaller numbers of large hay bales" instead of a needle in a haystack. The model is able to remove "background chatter," according to Dr. Johnson. The team has, since the model's creation, uncovered 196 pro-ISIS aggregates with more than 108k followers.
The lesson that larger intelligence agencies and the government can take is that the disbanding of smaller groups can help stop the growing momentum of a group.
Unfortunately, because social media is a big factor, the algorithm is incapable of weeding out lone attacks like that in Orlando and San Bernardino. However, in hindsight, the model could have predicted a large scale attack like the 2014 Siege of Kobani in Syria, which killed numerous civilians and displaced hundreds of thousands of people.