Using AI to investigate, scientists have developed a crime prediction algorithm with an accuracy of up to 90%

Minority Report is a very popular sci-fi movie about a special police unit that can catch murderers before they commit crimes with the help of three clairvoyant humans called clairvoyants. Precogs - can envision impending murders.

It's a great movie, if you're a fan of the sci-fi crime genre, or you're simply a fan of Tom Cruise. But the reason we're including it in this story is that a team of researchers just announced they've created an AI-powered system that can also predict crimes with 90% accuracy. And their system doesn't require the presence of precogs like in the movies, instead, "the system only needs past data and can predict future actions".

Advances in artificial intelligence (AI) and machine learning are of interest to governments, including the use of these tools to create predictive plans for police to prevent prevent crime.

Social and data scientists from the University of Chicago have developed an advanced algorithm that works by learning patterns of time and geographic location from publicly available data on violent crime and property. According to the University of Chicago, the new model can "isolate criminals by looking at the time and spatial coordinates of discrete events and then detecting patterns to predict future events. cities into spatial tiles more than 300 meters across and then predict crime in these areas rather than relying on neighborhoods or traditional political boundaries, which can also be biased."

The scientific report of the new study, published in the journal Nature Human Behavior on June 30, 2022, shows that the accuracy of the new algorithm's predictions is up to 90% and works not only for Chicago but also with data from seven other US cities: Atlanta, Austin, Detroit, Los Angeles, Philadelphia, Portland, and San Francisco.

"We've created a digital version of the urban environment. If you give it data from what happened in the past, it tells you what to expect in the future." Ishanu Chattopadhyay, PhD, Assistant Professor of Medicine at Chi University of Chicago and senior author of the study.

Data and social scientists at the University of Chicago have developed a new algorithm that predicts the state of security by machine learning the rules of time and geographic location from publicly available crime data. crime, property-related and violence. The algorithm proved successful in predicting future crimes a week in advance with an accuracy of about 90%.

This model has been tested and validated using historical data from Chicago on two major types of crime: violent crime (murder, assault, gangs) and property crime (theft) , robbery and theft of motor vehicles). And according to data obtained from the study, this AI-powered model can predict future crime a week in advance with about 90% accuracy.

This model and algorithm may not be perfect at the moment, but this AI-powered technology can certainly be a turning point in the global security industry.

Crime prediction models have been used by many law enforcement agencies before, but are not very accurate because they are based on a narrower set of factors, leaving out the nuanced relationship between crime, environment, and crime. the complex social environment of cities, and the impact of police enforcement.

• In 2012, the Chicago Police Department and academic researchers developed the "Risk Victim and Crime Model". The model uses factors such as age and arrest history to generate a list of potential shooters, or victims and perpetrators of mass shootings. This list later leaked out, causing many lawsuits. In 2017, an investigation by the Chicago Sun-Times revealed that nearly half of the people identified by the model as potential perpetrators had never been charged with illegal gun possession, of which 13% had never been charged. never been charged with a serious crime.