Burglary, as a prevalent and detrimental crime type, poses a major threat to public safety and property security. Accurate prediction of burglary occurrence is therefore critical. Although deep learning has achieved notable progress in crime prediction, the influence of varying urban spatial characteristics on predictive performance and resource efficiency remains underexplored. This study analyzes burglary prediction in eight representative cities from China, the United States, and Canada, uniformly employing the model based on convolutional neural networks and long short-term memory networks (CNN–LSTM) under a consistent spatio–temporal scale. A comprehensive evaluation framework based on precision, hit rate, prediction accuracy index (PAI), and prediction effectiveness index (PEI) was established, focussing on the validity of PEI and variations in optimal prediction thresholds across cities. Furthermore, standard deviation ellipses, Moran index, and kernel density analysis were applied to quantify spatial characteristics and explore their associations with PEI and resource allocation. The results indicate that cities with more concentrated spatial distributions and stronger spatial autocorrelation exhibit superior predictive efficiency and resource utilization. This study enriches the analytical scope of crime prediction efficiency and supports a shift from ‘accuracy–oriented’ to ‘efficiency–oriented’ modelling for intelligent allocation of urban public safety resources.

References

• Alves, L. G. A., Ribeiro, H. V., & Rodrigues, F. A. (2018). Crime prediction through urban metrics and statistical learning. Physica A: Statistical Mechanics and Its Applications, 505, 435–443. https://doi.org/10.1016/j.physa.2018.03.084
• Chainey, S., & Ratcliffe, J. (2005). Identifying crime hotspots. In Gis and crime mapping (pp. 145–182). John Wiley & Sons. https://doi.org/10.1002/9781118685181.ch6
• Chainey, S., Tompson, L., & Uhlig, S. (2008). The utility of hotspot mapping for predicting spatial patterns of crime. Security Journal, 21(1), 4–28. https://doi.org/10.1057/palgrave.sj.8350066
• Esquivel, N., Nicolis, O., Peralta, B., & Mateuet, J. (2020). Spatio-temporal prediction of Baltimore crime events using CLSTM neural networks. IEEE Access, 8,209101–209112. https://doi.org/10.1109/Access.6287639
• Gu, H., Chen, P., & Li, H. (2021). Overview and prospect for spatial-temporal prediction of crime (in Chinese). Journal of Geo-information Science, 23(1), 43-–57. https://doi.org/10.12082/dqxxkx.2021.200247
• He, R., Lu, Y., Jiang, C., Deng, Y., Li, X., & Shi, D. (2023). Progress in research and practice of spatial-temporal crime prediction over the past decade (in Chinese). Journal of Geo-information Science, 25(4), 866-–882. https://doi.org/10.12082/dqxxkx.2023.230299
• Hochreiter, S., & Schmidhuber, J. (1997). Long short-term memory. Neural Computation, 9(8), 1735–1780. https://doi.org/10.1162/neco.1997.9.8.1735
• Hu, T., Zhu, X., Duan, L., & Guo, W. (2018). Urban crime prediction based on spatio-temporal Bayesian model. PloS One, 13(10), e0206215. https://doi.org/10.1371/journal.pone.0206215
• Jefferson, M. (1939). The law of the primate city. Geographical Review, 29(2), 226–232. https://doi.org/10.2307/209944
• Jin, G., Wang, Q., Zhu, C., Feng, Y., Huang, J., & Zhou, J. (2020). Addressing crime situation forecasting task with temporal graph convolutional neural network approach. In 2020 12th International Conference on Measuring Technology and Mechatronics Automation (ICMTMA) (pp. 474–478). IEEE. https://doi.org/10.1109/ICMTMA50254.2020.00108
• Kianmehr, K., & Alhajj, R. (2006). Crime hot-spots prediction using support vector machine. In IEEE International Conference on Computer Systems and Applications (pp. 952–959). IEEE Computer Society. https://doi.org/10.1109/AICCSA.2006.205203
• Kianmehr, K., & Alhajj, R. (2008). Effectiveness of support vector machine for crime hot-spots prediction. Applied Artificial Intelligence, 22(5), 433–458. https://doi.org/10.1080/08839510802028405
• Law, J., Quick, M., & Chan, P. (2014). Bayesian spatio-temporal modeling for analysing local patterns of crime over time at the small-area level. Journal of Quantitative Criminology, 30(1), 57–78. https://doi.org/10.1007/s10940-013-9194-1
• Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (2002). Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11), 2278–2324. https://doi.org/10.1109/5.726791
• Lefever, D. W. (1926). Measuring geographic concentration by means of the standard deviational ellipse. American Journal of Sociology, 32(1), 88–94. https://doi.org/10.1086/214027
• Li, M., Du, W., Wen, S., Li, Q., Zhang, T., & Zhong, W. (2025). Crime forecasting: A spatio-temporal analysis with deep learning models. Journal of Computational Methods in Sciences and Engineering, 25(5), 4090–4099. https://doi.org/10.1177/14727978251337993
• Liu, L., Liu, W., Liao, W., Yu, H., Jiang, C., Lin, R., & Zhang, Z. (2018). Comparison of random forest algorithm and space-time kernel density mapping for crime hotspot prediction (in Chinese). Progress in Geography, 37(6), 761-–771. https://doi.org/10.18306/dlkxjz.2018.06.003
• Maheshwari, A., Camargo, M. E., Sathiyanarayanan, M., & Filho, W. P. (2025). Urban vs. rural crime dynamics: A spatio-temporal and machine learning approach for comparative analysis. In 2025 International Conference on Frontier Technologies and Solutions (ICFTS) (pp. 1–10). IEEE. https://doi.org/10.1109/ICFTS62006.2025.11031983
• Matereke, T., Nyirenda, C. N., & Ghaziasgar, M. (2021). A comparative evaluation of spatio-temporal deep learning techniques for crime prediction. In 2021 IEEE Africon (pp. 1–6). IEEE. https://doi.org/10.1109/AFRICON51333.2021.9570858
• McMillen, D. P. (2001). Nonparametric employment subcenter identification. Journal of Urban Economics, 50(3), 448–473. https://doi.org/10.1006/juec.2001.2228
• Mei, Y., & Li, F. (2019). Predictability comparison of three kinds of robbery crime events using LSTM. In Proceedings of the 2019 2nd International Conference on Data Storage and Data Engineering (pp. 22–26). Association for Computing Machinery. https://doi.org/10.1145/3354153.3354162
• Michael, T., Ross, H., & Janet, C. (2003). Infectious burglaries. A test of the near repeat hypothesis. The British Journal of Criminology, 43(3), 615–633. https://doi.org/10.1093/bjc/43.3.615
• Mohler, G. O., Short, M. B., & Brantingham, P. J. (2011). Self-exciting point process modeling of crime. Journal of the American Statistical Association, 106(493), 100–108. https://doi.org/10.1198/jasa.2011.ap09546
• Mohler, G. O., Short, M. B., & Brantingham, P. J. (2017). The concentration-dynamics tradeoff in crime hot spotting. In Unraveling the Crime-Place Connection (pp. 19–39). Routledge.
• Moran, P. A. P. (1950). Notes on continuous stochastic phenomena. Biometrika, 37(1/2), 17–23. https://doi.org/10.2307/2332142
• Parzen, E. (1962). On estimation of a probability density function and mode. The Annals of Mathematical Statistics, 33(3), 1065–1076. https://doi.org/10.1214/aoms/1177704472
• Pimple, S., Mansoor, H. D., & Karthikeyan, N. (2023). Enhanced public safety through LSTM-based spatio-temporal crime analytics. In 2023 International Conference on Emerging Research in Computational Science (ICERCS) (pp. 1–5). IEEE. https://doi.org/10.1109/ICERCS57948.2023.10434246
• Polvi, N., Looman, T., Humphries, C., & Pease, K. (1991). The time course of repeat burglary victimization. The British Journal of Criminology, 31(4), 411–414. https://doi.org/10.1093/oxfordjournals.bjc.a048138
• Ratcliffe, J. (2009). Crime mapping: Spatial and temporal challenges. In Handbook of quantitative criminology (pp. 5–24). Springer.
• Rosenblatt, M. (1956). Remarks on some nonparametric estimates of a density function. The Annals of Mathematical Statistics, 27(3), 832–837. https://doi.org/10.1214/aoms/1177728190
• Roshankar, R., & Keyvanpour, M. R. (2023). Spatio-temporal graph neural networks for accurate crime prediction. In 2023 13th International Conference on Computer and Knowledge Engineering (ICCKE) (pp. 168–173). IEEE. https://doi.org/10.1109/ICCKE60553.2023.10326223
• Rummens, A., & Hardyns, W. (2020). Comparison of near-repeat, machine learning and risk terrain modeling for making spatiotemporal predictions of crime. Applied Spatial Analysis and Policy, 13(4), 1035–1053. https://doi.org/10.1007/s12061-020-09339-2
• Rummens, A., Hardyns, W., & Pauwels, L. (2017). The use of predictive analysis in spatiotemporal crime forecasting: Building and testing a model in an urban context. Applied Geography, 86, 255–261. https://doi.org/10.1016/j.apgeog.2017.06.011
• Shen, H., Zhang, H., Zhang, Y., Zhang, Z., Zhu, Y., & Cai, L. (2019). Prediction of burglary crime based on LSTM (in Chinese). Journal of Statistics and Information, 34(11), 107–115. https://doi.org/10.3969/j.issn.1007-3116.2019.11.014
• Short, M. B., Mohler, G. O., Brantingham, P. J., & Tita, G. E. (2014). Gang rivalry dynamics via coupled point process networks. Discrete and Continuous Dynamical Systems – Series B, 19(5), 1459–1477. https://doi.org/10.3934/dcdsb.2014.19.1459
• Wang, S., Lin, Y., Jia, Y., Sun, J., & Yang, Z. (2024). Unveiling the multi-dimensional spatio-temporal fusion transformer (MDSTFT): A revolutionary deep learning framework for enhanced multi-variate time series forecasting. IEEE Access, 12,115895–115904. https://doi.org/10.1109/ACCESS.2024.3444788
• Zhang, Y., Yao, J., Li, L., & Geng, Z. (2023). Construction and application of crime prediction effectiveness index (in Chinese). Mathematics in Practice and Theory, 53(12), 163–175.
• Zhao, D., Du, P., Liu, T., & Ling, Z. (2023). Spatio-temporal distribution prediction model of urban theft by fusing graph autoencoder and GRU (in Chinese). Journal of Geo-Information Science, 25(7), 1448–1463. https://doi.org/10.12082/dqxxkx.2023.220584

To cite this article: Rui Wang , Yaofeng Zhang & Jinling Yao (2026) Optimal prediction threshold and spatial characteristics of burglary crime across cities, Statistical Theory and Related Fields, 10:2, 308-340, DOI: 10.1080/24754269.2026.2665854 To link to this article: https://doi.org/10.1080/24754269.2026.2665854