DSI professors mine social media to curb gang violence
A team of researchers from the Data Science Institute is studying how gangs use social media to communicate and recruit members. The researchers have developed tools to flag posts from gangs that can lead to violence. To prevent the posts from escalating into violence, they work with social workers who can intervene with gang members and defuse the anger that fuels their posts.
And now, thanks to a $548,000 grant from The Defense Advanced Research Projects Agency (DARPA), the research team will expand its work to study how ISIS uses social media to recruit members. DARPA is a U.S. Department of Defense agency that develops emerging technologies for the military. The research team will look at how gangs and ISIS use social media, and possibly build an aggression-indicator for DARPA that would help flagging aggressive posts by ISIS.
Desmond Patton, the lead on the project, said the project is "expected to have various applications, both for identifying indicators of ISIS recruitment on Twitter and curbing gang violence that's initiated on social media."
"We'll use an interdisciplinary approach that mixes qualitative analysis from the social sciences with latest techniques from computer science," added Patton, an affiliate of the Data Science Institute and professor of social work at Columbia University. "We'll compare differences and similarities between gang and ISIS recruitment with the goal of assessing how qualitative differences correlate to quantitative metrics and how qualitative insight can improve them."
Patton has collected millions of tweets from gang members in Chicago. He also intends to collect a similar amount of posts from ISIS. To help him sift through the ocean of data, he partnered with Kathleen McKeown, a former director of the Data Science Institute who is a renowned expert in natural language processing (NLP). A form of artificial intelligence, NLP trains computers to understand spoken and written language. It can filter through mountains of text and find patterns and themes. For this project, McKeown, a computer science professor at Columbia Engineering, will use NLP to flag aggressive social media posts from gangs and ISIS.
J.M. Berger, another team member, is a leading expert in social media analysis who will help to design metrics that measure in-group bonding in gangs and radicalization in ISIS. He's a fellow with George Washington University's Project on Extremism and has co-authored "The ISIS Twitter Census" and other landmark studies of how extremists use social media.
The team is not new to this work. For the last four years, Patton has studied what he refers to as Internet Banging — how gangs use social media to communicate and recruit. He has developed a qualitative approach to identifying offline and online indicators of gang violence called the Digital Urban Violence Analysis Approach.
He first took an interest in Internet Banging while working as a social worker in Chicago, when he came upon posts by Gakirah Barnes, a teenaged gangster with a murderous reputation. Nonetheless, Barnes had used Twitter to express a softer side, especially about a friend who was killed by police. Patton realized that online expressions of grief by gang members can be preludes to violence. Soon after those posts, Barnes was murdered by a rival gang member. He also realized that had he flagged her posts, social workers could have intervened and possibly saved Barnes's life and prevented a spiral of retaliatory violence. That's when Patton began collecting Twitter posts from gangs in Chicago. He now has millions of posts, and thanks to the DARPA grant, an opportunity to mine that data to deter violence.
"We mostly see gang members as bloody killers," said Patton. "But violence is complex and people are complex too. Gakira was violent but had a sensitive and humane side. Her posts gave us an opportunity to intervene and prevent her from being murdered and more blood being spilled. We missed that opportunity, but I hope the work we do for project will help us to detect and prevent violence that begins on social media."