Abstract When it comes to hot spot identification, spatial analysis techniques come to the fore. One of such techniques, that has gained great popularity among crime analysts, is the Kernel Density Estimation (KDE). Small variation in KDE parameters can give different outputs and hence affect predictive accuracy of hotspot map. The influence these parameters have on KDE hotspot output sparked many researches, mostly analyzing the influence of the cell size and bandwidth size. Yet, the influence of different classification methods applied to calculated cell values, including the choice of threshold value, on the KDE hotspot predictive accuracy remained neglected. The objective of this research was to assess the influence of different classification methods to KDE predictive accuracy. In each KDE computation, calculated cell values were divided into five thematic classes, using three the most common (default) classification methods provided by Environmental Systems Research Institute (ESRI) Geographical Information System (Arc GIS) (equal interval classification, quantile classification and natural breaks classification) and incremental multiples of the grid cells’ mean. Based upon calculated hit rates, predictive accuracy indices and recapture rate indices and taking into account the necessity that mapping output should satisfy some operational requirements as well as statistical rules, this research suggest that incremental mean approach with hotspot threshold of 3 and above multiples of the grid cell’s mean, should be used.

References

[1] Bowers K., Johnson S., and Pease K., “Prospective Hot-Spotting: The Future of Crime Mapping?” British Journal of Criminology, vol. 44, no. 5, pp. 641-658, 2004.

[2] Braga A. and Weisburd D., Policing Problem Places: Crime Hot Spots and Effective Prevention, Oxford: University Press, 2010.

[3] Caplan J. and Kennedy L., Risk Terrain Modeling: Crime Prediction and Risk Reduction, University of California Press, 2016.

[4] Chainey S., Reid S., and Stuart N., When is A Hotspot and Hotspot? A Procedure for Creating Statistically Robust Hotspot Maps of Crime, CRC Press, 2002.

[5] Chainey S. and Ratcliffe J., GIS and Crime Mapping, John Wiley and Sons, 2005.

[6] Chainey S., Tompson L., and Uhlig S., “The Utility of Hotspot Mapping For Predicting Spatial Patterns of Crime,” Security Journal, vol. 21, pp. 4-28, 2008.

[7] Chainey S., “Examining the Influence of Cell Size and Bandwidth Size on Kernel Density Estimation Crime Hotspot Maps for Predicting Spatial Patterns of Crime,” Bulletin of the Geographical Society of Liege, vol. 60, pp. 7-19, 2013.

[8] Eck J., Chainey S., Cameron J., Leitner M., and Wilson R., Mapping Crime: Understanding Hotspots, National Institute of Justice, 2005.

[9] Eldrandaly K. and Naguib S., “A Knowledge- Based System for GIS Software Selection,” The International Arab Journal of Information Technology, vol. 10, no. 2, pp. 152-159, 2013.

[10] Fan S., The Spatial-Temporal Prediction of Various Crime Types in Houston, TX Based on Hot-Spot Techniques, Thesis, Louisiana State University, 2014.

[11] Ghosh A., Lagenbacher M., Duda J., and Klofas J., “The Geography of Crime in Rochester- Patterns over Time (2005-2011),” Working Paper, Center for Public Safety Initiative, 2012.

[12] Haberman C. and Ratcliffe J., “The Predictive Policing Challenges of Near Repeat Armed Street Robberies,” Policing: A Journal of Policy and Practice, vol. 6, no. 2, pp. 151-166, 2012.

[13] Harries K., Mapping Crime: Principle and Practice, Office of Justice Programs, 1999.

[14] Hart T. and Zandbergen P., “Effects of Data Quality on Predictive Hotspot Mapping,” Technical Report, Submitted to US Department of Justice, 2012.

[15] Hart T. and Zandbergen P., “Kernel Density Estimation and Hotspot Mapping: Examining the Influence of Interpolation Method, Grid Cell Size, And Bandwidth on Crime Forecasting,” Policing: An International Journal of Police Strategies and Management, vol. 37, no. 2, pp. 305-323, 2014.

[16] Johnson S., Bowers K., Birks D., and Pease K., “Predictive Mapping: Accuracy, Units of Analysis and the Environmental Backcloth,” in Proceedings of Putting Crime in its Place: Units of Analysis in Geographical Criminology, New York, pp. 171-198, 2009.

[17] Johnson S., Bernasco W., Bowers K., Elffers H., Ratcliffe J., Rengert G., and Townsley M., “Space-Time Patterns of Risk: A Cross National Assessment of Residential Burglary 1062 The International Arab Journal of Information Technology, Vol. 16, No. 6, November 2019 Victimization,” Journal of Quantitative Criminology, vol. 23, no. 3, pp. 201-219, 2007.

[18] Levine N., “The “Hottest” Part of A Hotspot: Comments on “The Utility of Hotspot Mapping for Predicting Spatial Patterns of Crime,” Security Journal, vol. 21, no. 4, pp. 295-302, 2008.

[19] Levine N., CrimeStat: A Spatial Statistics Program for the Analysis of Crime Incident Locations (v 4.0). Ned Levine and Associates, Houston, TX and the National Institute of Justice, 2013.

[20] Monmonier M., “Lying with Maps,” Statistical Science, vol. 20, no. 3, pp. 215-222, 2005.

[21] Ratcliffe J. and Rengert G., “Near Repeat Patterns in Philadelphia Shootings,” Security Journal, vol. pp. 21, 58-76, 2008.

[22] Sherman L., Hot Spots of Crime and Criminal Careers of Places, Criminal Justice Press, 1995.

[23] Smith S. and Bruce C., CrimeStat III: User Workbook, the National Institute of Justice, 2008.

[24] Tompson L., Partridge H., and Shepherd N., “Hot Routes: Developing a New Technique for the Spatial Analysis of Crime,” Crime Mapping: A Journal of Research and Practice, vol. 1, no. 1, pp. 77-96, 2009.

[25] Van Patten I., McKeldin-Conor J., and Cox D., “A Microspatial Analysis of Robbery: Prospective Hot Spotting in A Small City,” Crime Mapping: A Journal of Research and Practice, vol. 1, no. 1, pp. 7-32, 2009.

[26] Walter L., McInnis B., Price C., Smith S., and Hollywood J., Predictive Policing: The Role of Crime Forecasting in Law Enforcement Operations, RAND Corporation, 2013.

[27] Williamson D., McLafferty S., Goldsmith, V., Mollenkopf, J., and McGuire P., “A Better Method to Smooth Crime Incident Data,” ESRI Arc User Magazine, pp. 1-5, 1999.

[28] Zandbergen P., “Geocoding Quality and Implications for Spatial Analysis,” Geography Compass, vol. 3, no. 2, pp. 647-680, 2009. Nenad Milić received the B.Sc. degree and M.Sc. degree in 2000. and 2004. Respectively from the Police Academy, Belgrade, Serbia, and Ph.D. degree in 2012. from the Faculty of Law, University of Kragujevac, Serbia. He is Associate Professor at the Academy for Criminalistic and Police Studies in Belgrade, Serbia, Department of Criminalistic. His research interest includes GIS, crime mapping and crime analysis. Brankica Popović received the B.Sc. degree in electrical engineering in 1990. from the Faculty of Electrical Engineering, University of Pristina, M.Sc. degree in 1997. from the Faculty of Electrical Engineering, University of Belgrade and the Ph.D. degree in 2011 from the Faculty of Electronic Engineering, University of Nis, Serbia. She is Associate Professor at the Academy for Criminalistic and Police Studies in Belgrade, Serbia, Department of Informatics and Computer Sciences. Her research interest include image processing, biometric systems, information security and information systems with emphasize on information technology utilization in law enforcement agencies. She is a member of IEEE. Saša Mijalković received the B.Sc. degree at Police Academy (2001), M.Sc. degree (2004) and the Ph.D. degree (2006) from the Faculty of Security of the University of Belgrade, Serbia. He is Full Professor at the Academy for Criminalistic and Police Studies in Belgrade, Department of Criminalistic. His research interest includes national security, organized crime, terrorism, intelligence and security affairs and services. Darko Marinković received the B.Sc. degree in 1999. from the Police Academy, Belgrade, Serbia, M.Sc. degree and Ph.D. degree in 2003. and 2008. respectively from the Faculty of Law, University of Belgrade, Serbia. He is Full Professor at the Academy for Criminalistic and Police Studies in Belgrade, Serbia, Department of Criminalistic. His research interest include criminal investigation, organized crime, drug related crimes, special investigative techniques.