The Los Angeles Police Department is using a new tactic in their fight against crime called “predictive policing.” It's a computer program that was originally developed by a team at UCLA, including mathematician Andrea Bertozzi and anthropologist Jeff Brantingham. “Science Behind the News” is produced in partnership with the National Science Foundation.
Science Behind the News- Predictive Policing
ANNE THOMPSON, reporting:
Los Angeles, California, second most populous city in the United States, spread out over 468 square miles- with nearly 4 million residents and fewer than 10,000 police officers sworn to protect and serve, keeping order is a full time job.
Captain SEAN MALINOWSKI (Los Angeles Police Department, Foothill Division): We are playing probabilities, and putting officers in the right place at the right time.
THOMPSON: To make the most of their resources, police captains like Sean Malinowski of the LAPD, and other police departments across the country, are turning to a new tactic to help stop crime before it happens. It's called predictive policing.
MALINOWSKI: It calculates for the next 12 hours in the future what areas have the highest probability of a crime occurring. And then what we do, we provide that information to the officers and then the officers go out and they try to prevent those crimes from occurring.
THOMPSPN: The predictive policing program used by the LAPD didn't come out of a police academy; it's based on research by a team of mathematicians and social scientists at UCLA trying to predict where and when crime is most likely to happen.
Prof. JEFF BRANTINGHAM (University of California, Los Angeles): I had these ideas that really, human behavior is actually quite predictable and that you can study human behavior and understand where crime patterns come from in a very quantitative way.
THOMPSPN: With funding from the National Science Foundation, anthropology professor Jeff Brantingham teamed up with mathematics professor Andrea Bertozzi and others to analyze crime patterns and develop computer models to simulate criminal behavior. Their crime prediction model is based on the same algorithms used to predict earthquakes and aftershocks.
Prof. ANDREA BERTOZZI (University of California, Los Angeles): Once an event happens, it triggers another event. And so you can apply this idea to gang crimes, you can apply it to burglaries, you can apply to automobile theft. There are many types of activities for which this idea is very relevant.
THOMPSON: Thousands of pieces of crime data from the LAPD, including the locations, times, and dates of past crimes, are processed by the software program, known as PredPol, to calculate and predict the potential criminal activity for an area at a certain time.
BRANTINGHAM: And those predictions are delivered back to the police departments in a way that allows them to use it in real time fashion.
THOMPSON: At roll call in the Foothill Area of Los Angeles, where UCLA tested the program in 2011…
LT. JULIE S. RODRIGUEZ (Los Angeles Police Department, Foothill Division): Ok, so today we’re going to hand out the predictive policing maps.
THOMPSON: …officers receive maps showing the areas of predicted activity for the next shift.
Red boxes on the map highlight the hotspots, areas measuring 500 feet by 500 feet that will require extra patrols.
MALINOWSKI: So the officers know that’s the highest probability area where they should be looking for a crime to be committed. And we ask them to get in there and disrupt the crime from occurring or deny the criminal the opportunity to commit the crime.
THOMPSON: That's just what happened in 2011, in Santa Cruz, California, where officers were patrolling a hotspot - putting them in the right place at the right time to stop an assault.
SERGEANT CHRISTIAN LA MOSS (file): We came here, did extra patrols and we were…
PETER ALEXANDER (file): Stopped a crime.
LA MOSS (file): We were able to stop a crime in progress before it got worse.
BERTOZZI: We are helping police fight crime by giving them the best state of the art mathematical models and algorithms to take the data from yesterday and today and figure out what’s going to happen tomorrow in the field.
THOMPSON: With mathematics and social sciences, the police have a new weapon in their arsenal, helping to not only protect and serve, but also to predict a crime before it happens.
Rob Mastrodomenico has just got back from a lunchtime kickabout (an informal game of soccer) with his colleagues. "I'm still a bit out of breath," he says apologetically, despite looking considerably less out of breath than I do after the short walk from Kentish Town tube station. I assume they've all been playing on nearby Hampstead Heath but his reply is unexpected: "No, in a client's back garden."
Science Behind the News, Crime, Science, Math, Mathematics, Predictive Policing, Police, Criminals, Criminal, Behavior, Analysis, Computer, Computer Program, Models, Algorithms, Probability, Earthquake, Prediction, PredPol, Los Angeles Police Department, LAPD, Sean Malinowski, Andrea Bertozzi, Jeff Brantingham, University of California, Los Angeles, UCLA, California, Santa Cruz, National Science Foundation, NSF