Software used to predict the killer instinct

By on
Software used to predict the killer instinct

Criminologist promises statistical likelihood of an individual becoming a
murderer.

Criminologist promises statistical likelihood of an individual becoming a
murderer.

An American statistician is developing software which he hopes will predict the likelihood of someone committing murder.

Richard Berk, a criminologist at the University of Pennsylvania, is using data from a local police department to build the application, which will examine 30-40 data sets and come up with a statistical likelihood of an individual becoming a killer.

The first version of the software will be released in the New Year and trials will start in the spring.

"You can imagine the indicators that might incline someone towards violence, such as youth, having committed a serious crime at an early age, being a man rather than a woman, and so on," Berk told The Kansas City Star.

"Each, by itself, probably isn't going to make a person pull the trigger. But put them all together and you've got a perfect storm of forces for violence."

The software uses two years' police data to formulate a model, and is designed to provide social workers with information on where to target their resources.

Berk explained that one of the most common indicators of murderous intent is an early exposure to, or propensity for, violence with youth being another factor. Once a person reaches 30 the likelihood that they will commit a murder is sharply reduced.

"If we have 100 probationers I can accurately find the one murderer who will statistically be in that group if I devote resources to all 100 as if they are murderers," he said.

"The problem is that for that one murderer who is a 'true positive' I have 99 false positives. We all would agree that this not a good use of resources.

"Now suppose I can identify the 10 at highest risk, for that one true positive I now have nine false positives, and that may be something we choose to live with."
Copyright ©v3.co.uk
Tags:

Most Read Articles

Log In

Username:
Password:
|  Forgot your password?