US researchers have developed a statistical model that they say predicts a budding terror group’s future lethality based on its first 10 to 20 attacks.
The model could help security forces to pick out and focus on more deadly terror groups before they carry out more serious attacks.
It uses publicly available data from the Global Terror Database and the RAND Worldwide Terrorism Incidents data compilation site.
The researchers from Northwestern University scoured data on terror groups that were active from 1970 to 2014, the university said in a statement.
The model is based on systems that predict the success of young businesses.
“Essentially we said, ‘What if we think of terror organizations like a business whose product is lethality? How do we predict their success in producing that product?’” Brian Uzzi, one of the study’s authors, said in a statement.
Business investors look at publicly available information to extrapolate a company’s future success, but since such information on terror groups is not available, the researchers looked for alternatives to use as proxies.
Investors pay close attention to the timing of product releases, for example. A company that releases new products at regular intervals likely has more resources than one that launches products randomly, so as a proxy, the researchers looked at the cadence of attacks as an indication of a group’s level of organization and resources.
They also looked at the arsenal the groups used in attacks, the sophistication of their weaponry, and how successful they were in their attacks.
The Islamic State, for example, at the beginning carried out attacks irregularly, but the lethality of those attacks was near the 90th percentile relative to other terror groups. This ranked the group as potentially very deadly after its first 10 attacks, the researchers said.
The data accurately predicted that some groups that carried out few attacks in their early years would become much more deadly later on, including Al-Shabaab in east Africa.
“The model can predict the future impact of some of these sleeper groups even while they are still operating in an under-the-radar way,” said Yang Yang, one of the study’s authors.
Looking at the first 10 attacks, the model is about 60 percent predictive when placed next to a group’s lifetime data, the researchers said.
The US spends $500 million annually in the fight against terror, and from 2000 to 2015, an average of 61 new terror groups appeared per year, the researchers said, citing statistics by the Global Terror Database.
“This early warning is huge because not only can it help the government target and neutralize the groups with the most potential for destruction, it also can help the government strategically deploy resources and avoid spending billions of dollars fighting a group that is likely to burn out on its own anyway,” Uzzi said.
The study by data scientists from Northwestern University’s Kellogg School of Management, titled “Quantifying the Future Lethality of Terror Organizations,” was published on October 7 in the PNAS science journal and funded by the US military.