New Predictive Policing Algorithm could Predict Crime Hotspots in Real-Time

With police departments around the world stretching their budgets thinner almost every year, research in the area of predictive policing has been intensifying to meet the rising need for more complex software.

A new algorithm developed by the University of Surrey and Georgia Institute of Technology could finally give authorities the upper hand, thanks to its ability to quickly process real-time data and predict where illegal activity could reoccur.

In a paper out in the journal Computational Statistics and Data Analysis, the research team details their approach, which combines the Epidemic Type Aftershock Sequence (or ETAS) – a popular, grid-based method for modelling urban crime – with techniques more frequently seen in weather forecasting and space missions.

The new algorithm, called the Ensemble Poisson Kalman Filter (or EnPKF) can provide real-time forecasts of the crime rate and even give an indication of how likely a crime is to reoccur in a specific location, or point out where a temporary crime hot-spot could arise.

Ready to give it a spin, the research team deployed the algorithm on a data-set comprised of data on more than 1,000 violent crimes in Los Angeles, committed by 33 known gangs between the years 1999 through 2002.

Combining a method which outperforms human analysts with techniques used in weather forecasting, the new algorithm could monitor crime levels in real-time. Image credit: Puamelia via flickr.com, CC BY-SA 2.0

“We are cautiously excited about the Ensemble Poisson Kalman Filter, an approach that has given us an insight into when crime can be predicted, and has shown us the importance of using real-time data to make the overall system stronger,” said Dr David Lloyd from the University of Surrey’s Department of Mathematics.

According to Lloyd, the research team is already well underway to boosting the efficiency of their new tool and have even tested it against data from Chicago.

“It is important to remember that EnPKF, and algorithms similar to this, are tools used to help our law enforcement who work hard to keep our communities safe. Their use will ultimately be determined by the needs of individual departments.”

Apart from predictive policing, the algorithm could also be used to monitor train delays, earthquake aftershocks and even insurance claims in sub-Saharan Africa.

Source: phys.org.


<!–

Comment this news or article

–>