Predictive policing has become a widely discussed phrase in the last few years, sometimes being used so broadly and for so many technologies that the real performance features and technical specifications get lost. This post seeks to succinctly establish parameters for a current definition of what predictive policing is.
Predictive Policing Definition: NIJ
Predictive policing tries to harness the power of information, geospatial technologies and evidence-based intervention models to reduce crime and improve public safety. This two-pronged approach — applying advanced analytics to various data sets, in conjunction with intervention models — can move law enforcement from reacting to crimes into the realm of predicting what and where something is likely to happen and deploying resources accordingly.
The National Institute of Justice, the research, development and evaluation agency of the United States Department of Justice, notes that law enforcement work is frequently reactive: officers respond to calls for service, quell disturbances and make arrests. In contrast, notes the NIJ, proactive policing uses data, geospatial models, and intervention models to predict what and where [and when] something is likely to happen and to deploy resources accordingly.
Very importantly, the NIJ explains, “The predictive policing approach does not replace traditional policing. Instead, it enhances existing approaches such as problem-oriented policing, community policing, intelligence-led policing and hot spot policing.”
Definition Parameter: Predictive Policing is not “Hotspot” mapping of past crimes
While crime may afflict the same neighborhoods year after year, the day-to-day fluctuations in where and when crimes occur are large. Extensive research has shown that day-to-day crime patterns are the result of:
- Crime generators that are fixed features of the environment
- Repeat and near-repeat victimization that leads previous victims and their neighbors to be at greater risk of follow-on crimes
- The routine activity patterns of offenders that keep risk local
- substantial random noise.
Each of these processes is well known empirically, but when put together, their impact on how crime hotspots emerge, spread, and disappear is incredibly complex. This makes predictions for where crime will occur in the next 10-12 hours very difficult.
Predictive Policing Definition: RAND Corp.
The application of analytical techniques, particularly quantitative techniques, to identify promising targets for police intervention with the goal of reducing crime risk by preventing future crimes or solving past crimes. (source)
Developments in mathematical and statistical modeling, high-performance cloud computing, and GPS-enabled mobile devices now make it possible for frequently updated crime forecasts to be at the disposal of officers in the field. Knowledge, skills, and experience are essential and indispensable. They can reliably direct officers to the top two or three riskiest locations in their operational environment for any particular shift on any particular day. It is much harder and time-consuming, though, for law enforcement to identify and choose between a few dozen locations where risk might be lower and is highly variable from day to day.
Opportunities to disrupt crime in these places are missed and thus crime prevention and reduction becomes unnecessarily difficult. Chronic hotspots are long term in duration and necessitate problem-oriented policing strategies to address the root causes of crime. Temporary hotspots, on the other hand, last on the time scale of days to weeks.
Definition Parameter: Predictive Policing does not replace law enforcement veterans’ intuition or experience
Using high-powered mathematics and near real time crime data housed in a department’s RMS, yesterday’s crime can be evaluated in the context of all crimes occurring over a long horizon and wide spatial fields to calculate accurate probabilities for where and when crime will occur today.
Definition Parameter: Predictive Policing is not a faster reaction to ongoing crime, but a more accurate forecast of future events
Within these parameters, predictive policing can help law enforcement agencies intelligently inform patrol officers of where and when crime may occur, allowing them to effectively prevent crime. A minimal amount of data is needed to go beyond traditional “hotspot” mapping and define predictive policing.
More About Field-Tested Predictive Policing with PredPol:
- PredPol needs only three pieces of data: type of crime, place of crime, and time of crime. There is no personal data or confidential information, thus avoiding any privacy and profiling risks, while obviating the need for PredPol access to a department’s more sensitive information.
- Newer events are weighted heavier than older ones so that predictions more effectively reflect near term events.
- PredPol’s technology has been helping law enforcement agencies to dramatically reduce crime in jurisdictions of all types and sizes, across the U.S. and overseas.