Prediction-Led Policing Essay

Predictive policing, mass data-analysis for crime prediction

Abstract. Predictive policing is the application of data-analysis on crime data to predict criminal activity. This short paper presents an overview of the prediction-led policing process.

1. Introduction. The effectiveness of the police is limited to their available resources. Smart and anticipatory policing is necessary to combat crime better and to use limited resources more efficiently. Predicting future crime has recently gained notable currency in law enforcement. By finding patterns in data on past crimes the police can identify potential future criminality. This short paper surveys the different steps taken in this predictive-led policing.

2. Preliminaries. We assume familiarity with the basic concepts of regression analysis [1], and

Most police departments collect crime data. To make predictions however, you also need information on the environment these crimes took place in. The various data sources need to be combined by analysts. Something to keep in mind is that predictions heavily depend on the amount and quality of the data. If it lacks one of these, the predictions become unreliable.

Once the data is collected, it has to be analysed by analytical techniques; the second step. Various techniques like regression and data-mining can be used on the data to find crime patterns. The resulting patterns can then be used to inform a method that identifies hot spots. As important it is to have good data, it is also important to have good analytical techniques. Poor quality techniques will result in poor predictions.

The third step is the intervention of the police based on the predictions. The type of intervention vary according to the problem. If there is an increased risk of local crimes, the police can relocate more resources to that area. Predictive policing also allows the police to identify the sources of crime and combat them