Advertisement

New model predicts characteristics of terror attacks with 90 percent accuracy

"Predicting terrorist events is a dream, but protecting some area by using patterns is a reality," said researcher Salih Tutun.

By Brooks Hays
A new predictive framework designed to anticipate the characteristics of future terrorist attacks could help police and security forces better prepare and mitigate risks. Photo by Maya Vidon-White/UPI
A new predictive framework designed to anticipate the characteristics of future terrorist attacks could help police and security forces better prepare and mitigate risks. Photo by Maya Vidon-White/UPI | License Photo

March 2 (UPI) -- Researchers have developed a new model to predict the characteristics of future terrorist behaviors based on past attacks.

The new model is called Networked Pattern Recognition Framework, or NEPAR. It was designed by systems engineers at Binghamton University, State University of New York.

Advertisement

Researchers populated their model with the specifics of 150,000 terrorist attacks carried out between 1970 and 2015. Specifics include factors like attack time and weapon type. The framework analyzes relations among the dataset, tracing patterns and ultimately predicting the characteristics of future attacks.

The model's accuracy can be tested by comparing its predictions with the specifics of real attacks. Tests showed the framework can predict attack characteristics with a 90 percent accuracy.

"Terrorists are learning, but they don't know they are learning," researcher Salih Tutun, a doctoral student at Binghamton, said in a news release. "If we can't monitor them through social media or other technologies, we need to understand the patterns. Our framework works to define which metrics are important."

The framework operates by defining important qualities and then locating similarities among other terror attacks, which researchers call the interaction function. The algorithm then looks for key differences among the interactions.

Advertisement

"For example, what is the relationship between the Paris and the 9/11 attacks? When we look at that, if there's a relationship, we're making a network," Tutun said. "Maybe one attack in the past and another attack have a big relationship, but nobody knows. We tried to extract this information."

Most terrorism analysis has focused on the behavioral patterns of individual terrorists. The latest study -- detailed in the journal Expert Systems with Applications -- is one of the first to focus on patterns and relationships among the attacks themselves.

Tutun said the research isn't about predicting exactly when, where and by who the next terrorist attack will be executed. Instead, it's about analyzing probabilities and identifying risks -- risks that can be mitigated.

"Predicting terrorist events is a dream, but protecting some area by using patterns is a reality. If you know the patterns, you can reduce the risks. It's not about predicting, it's about understanding," said Tutun.

Latest Headlines