Department of Justice awards $731,000 to Brandeis University and Colorado State University researchers to create a tool to predict terrorist radicalization

After a terrorist attack, the question always arises: Could this have been prevented?

The answer may lie at the intersection of data science and social science.

The National Institute of Justice, the research arm of the U.S. Department of Justice, has awarded Brandeis politics professor Jytte Klausen and Colorado State University professor of electrical and computer engineering Anura Jayasumana a joint grant of $731,000 to develop a data-driven, risk assessment protocol that can help law enforcement identify individuals that are on the path to radicalization, focusing on overt behaviors indicative of extremism rather than country of origin or religion.

“The collaboration marries cutting edge computational techniques and old-fashioned social science methodologies to analyze how people are drawn into militant terrorist networks,” Klausen said. “We aim to use pattern-detection computational methods to analyze a large real-world data collection tracking the behavioral characteristics of known terrorist offenders that we have compiled from court documents and the offenders’ own social media.”

The research will help a number of key groups – from law enforcement to social workers and probation officers working with violent extremists – to better assess risk. The collaboration also marks a break-through in the use of second-generation (2G) computational techniques in social science.  

"There is no protocol to identify, in real time, people getting radicalized," said Jayasumana. "On top of that, law enforcement does not have the resources to look at millions of people…Our purpose is to use the characteristics identified as radicalization indicators and narrow down at-risk groups for law enforcement to examine."

Jayasumana, a leader in network science with an expertise in data sciences related to pattern and anomaly detection, network mapping for IoT (Internet of Things), and communication among weather radars, will create an algorithm to power the tool built on research from Klausen’s Western Jihadism Project.

Launched in 2009, the Western Jihadism Project is a multimedia data archive that records the growth of Jihadism in Western Europe, North America and Australia since the early 1990s.

The database comprises records of thousands of terrorism offenders from 20 countries, plus information about the personal social networks in Jihadist terrorist organizations and their recruitment efforts in Western nations. Brandeis’ research team at the Western Jihadism Project – which includes Klausen and a staff of undergraduate and graduate students – will continue collecting data about subjects who commit acts of terror and share it with Jayasumana’s team of computer scientists at Colorado State.

“This project relies on research collected only from publicly available data and refocuses counter-terrorism efforts by examining risk behaviors rather than making assumptions based on demographics,” Klausen said. “By creating an evidence-based approach that helps us understand how people become dangerously radicalized, we can take steps to ensure public safety without profiling innocent people or threatening their privacy and civil liberties.”