AI model that predicts crime in US cities is right nine times out of ten

A AI model correctly predicted crimes a week before they happened with 90% accuracy in eight US cities, the co-creator of the model said.

Ishanu Chattopadhyayan assistant professor at the University of Chicago, told Insider that from 2014 to late 2016, he and his team created an “urban twin” to monitor crime data in Chicago before predicting the probability of certain crimes for the following weeks, with 90% accuracy within two blocks.

The model, which had similar results in seven other cities, focused on the types of crimes committed and where they took place. Chicago crime rates in 2020 were 67% higher than the national average, according to data collected by AreaVibes

Racial prejudice in the police has high economic costs and contributes to inequality in areas already suffering from high deprivation, according to research compiled by Econofact

While some models try to eradicate these biases, they often have: had the opposite effectwith allegations that racial bias in the underlying data amplifies future biased behavior.

In 2016, the Chicago Police Department tried out a model to predict those most at risk of being involved in a shooting, but the secret list finally revealed that 56% of black men who lived in the city appeared on the listincitement to accusations of racism.

Chattopadhyay said their model found that arrests increased alongside reported crime in high-income neighborhoods, while arrests were flat in lower-income areas, suggesting some bias in police response.

“We demonstrate the importance of discovering city-specific patterns for the prediction of reported crime, which generates fresh perspectives on urban neighborhoods, enables us to ask new questions and allows us to evaluate policing in new ways. co-author James Evans told Science Daily

Lawrence Sherman of the Cambridge Center for Evidence-Based Policing told the New Scientist he was concerned about the inclusion of police data in the investigation that depended on reporting by citizens or the crimes that the police are looking for.

Chattopadhyay agreed that this was a problem, and that his team had tried to explain it by excluding citizen-reported crimes and police interventions, usually involving minor drug offenses and traffic congestion, and zoning them into more serious violent crimes. and property crimes that were more likely to be reported in any setting.

Chattopadhyay, who made the data and algorithm public to increase scrutiny, hoped the findings would be used for high-level policy and not as a reactive tool for police.

“Ideally, if you can predict or prevent crime, the only response is not to send more officers or flood a particular community with law enforcement,” Chattopadhyay said. “If you could prevent crime, there are plenty of other things we can do to prevent things like this from actually happening so that no one goes to jail and helps communities as a whole.”

Leave a Comment

Your email address will not be published.