The algorithm is able to predict crimes a week ahead

The algorithm was developed by University of Chicago experts who used historical data on violent crime (murder, assault, and injury) and property crime (burglary, theft, and auto theft) in the city to test and validate the model.

These crimes were chosen because they are less likely to be subject to the kind of law enforcement bias often present in drug-related and similar crimes. Of course, if you’ve seen the minority report, this sounds like something to you.

Dividing cities into spatial mosaics, the algorithm is able to detect patterns and try to predict future crimes, which it does with 90% accuracy.This is stated in a study published in the journal Nature Human Behavior.

And it wasn’t just in Chicago that the algorithm had an almost infallible ability to predict crime. The system worked just as well after connecting data for other US cities: Atlanta, Austin, Detroit, Los Angeles, Philadelphia, Portland and San Francisco, they explain.

The press release states that, unlike other future crime detection tools, the algorithm does not describe crime as originating from hot spots that spread to nearby areas.

This method can ignore the complex social environment of cities. and the relationship between crime and the consequences of policing.

Lead author Ishanu Chattopadhyay warned that the tool should not be used to manage police forces despite its accuracy. That is, departments should not use it for proactive area patrols, such as crime prevention.

Earlier this week, we learned that Chinese regulators wanted to use various data collected about Chinese citizens. create profiles by which an automated system could predict potential dissidents or criminals.

It’s another step in that direction, only it seems less intimidating than China’s system for detecting crimes before they happen, making it something of a Minority Report-style system.