Comparison of near-Repeat, Machine Learning and Risk Terrain Modeling for Making Spatiotemporal Predictions of Crime
- PDF / 533,109 Bytes
- 19 Pages / 439.37 x 666.142 pts Page_size
- 86 Downloads / 175 Views
Comparison of near-Repeat, Machine Learning and Risk Terrain Modeling for Making Spatiotemporal Predictions of Crime Anneleen Rummens 1
& Wim
Hardyns 1
Received: 8 January 2020 / Accepted: 22 March 2020/ # Springer Nature B.V. 2020
Abstract The main objective of this study is to test and compare the prediction performance of three of the most common predictive policing methods. A near-repeat model, a supervised machine learning model, and a risk terrain model are tested and compared against each other using retrospective analysis of home burglary crime data from a Belgian city. Hotspot analysis is included as a baseline. Predictions are made for three different months (January, May and September 2017) to account for seasonal differences. Variations in spatial context (city center vs. suburbs) and the number of predicted risk locations are also tested. Prediction performance is measured using accuracy, nearhit rate, precision and F1-score. The results show that there are some notable differences in prediction performance between the model types across the tested variations. In general, the ensemble model tends to be the most consistent high performer across all tested variations. Also notable is that hotspot analysis is not clearly outperformed by the other methods. The different methods have their own strengths and weaknesses and optimal prediction performance crucially depends on the specific location context. More comparative analyses of predictive policing methods in different contexts are needed to gain a more complete picture. Future research could also focus on how combining methods can help improve crime prediction performance. Keywords Predictive policing . Near-repeat . Machine learning . Risk terrain modeling .
Hotspot analysis
* Anneleen Rummens [email protected]
1
Institute for International Research on Criminal Policy (IRCP), Department of Criminology, Criminal Law and Social Law, Ghent University, Universiteitstraat 4, B-9000 Ghent, Belgium
A. Rummens, W. Hardyns
Introduction Decision-making processes in policing are increasingly guided by intelligence gained from predictive analysis (Ratcliffe 2016). In the context of crime data analysis, predictive analysis methods (commonly called predictive policing in this context) are used to make spatiotemporal predictions of criminal events. Predictive policing can be defined as: “the use of historical data to create a spatiotemporal forecast of areas of criminality or crime hot spots that will be the basis for police resource allocation decisions with the expectation that having officers at the proposed place and time will deter or detect criminal activity” (Ratcliffe 2014, p. 4). Predictive policing specifically aims to improve the specificity of crime forecast compared to traditional methods such as hotspot analysis. The development of predictive policing interacts with other developments in criminology, in particular the use of big data and predictive modeling (Chan and Moses 2015; Moses and Chan 2016), combined with an increasing use of
Data Loading...
