Spatio-temporal cokriging crime predictions using social media data: a multi-type case study in San Jose, California

Abstract

Crime prevention requires accurate prediction of the spatial and temporal distribution of criminal activities to effectively allocate law enforcement resources. However, many trending crime prediction algorithms lack comprehensive spatio-temporal structures and often consider only single input variables. This study innovatively using in ST-Cokriging method integrated both historical crime records as the primary variable and crime-related geo-tagged Twitter data as the co-variable for crime prediction. The predictive method has been specifically developed to assess crime risk across three major crime types—street crime, property crime, and vehicle crime—and applied in the San Francisco Bay Area (SFBA), California, a region characterized by high development and heightened crime sensitivity, for both prediction and validation. The results indicate that incorporating social media data into a spatio-temporal statistical method improves the associations between predicted and actual crime risk, reduced the Root Mean Squared Error (RMSE), and enhanced the identification of crime risk areas for both weekdays and weekends across three crime types compared to the method without the co-variable. This study presents a new multi-variable approach to more accurately predict crime, enabling law enforcement proactively address crime of varying nature in urban areas.

Keywords

Public Safety / GIS / Spatio-temporal Analysis / ST-Cokriging / Social Media / Urban Systems

References

[1] Ahn, H. Il, & Spangler, W. S. (2014). Sales prediction with social media analysis. Annual SRII Global Conference, SRII, Annual SRII Global Conference (SRII), 213–222. https://doi.org/10.1109/SRII.2014.37

[2] Alves LGA, Ribeiro HV, Rodrigues FA. Crime prediction through urban metrics and statistical learning. Physica a: Statistical Mechanics and Its Applications, 2018, 505: 435-443

[3] Amerio P, Roccato M. A predictive model for psychological reactions to crime in Italy: An analysis of fear of crime and concern about crime as a social problem. Journal of Community & Applied Social Psychology, 2005, 15(1): 17-28

[4] Bendler, J., Brandt, T., Wagner, S., & Neumann, D. (2014). Investigating crime-to-twitter relationships in urban environments - Facilitating a virtual neighborhood watch. ECIS 2014 Proceedings - 22nd European Conference on Information Systems, July 2017. https://www.wi.uni-muenster.de/research/publications/169230

[5] Berry-James RJM, Gooden ST, Johnson RG. Civil rights, social equity, and census 2020. Public Administration Review, 2020, 80(6): 1100-1108

[6] Braga AA, Papachristos AV, Hureau DM. The effects of hot spots policing on crime: An updated systematic review and meta-analysis. Justice Quarterly, 2014, 31(4): 633-663

[7] Butt, U. M., Letchmunan, S., Ali, M., & Sherazi, H. H. R. (2025). START: A Spatiotemporal Autoregressive Transformer for Enhancing Crime Prediction Accuracy. IEEE Transactions on Computational Social Systems. https://doi.org/10.1109/TCSS.2025.3550196

[8] Chainey S. Examining the influence of cell size and bandwidth size on kernel density estimation crime hotspot maps for predicting spatial patterns of crime. Bsglg, 2013, 60(1): 7-19

[9] Chainey S, Tompson L, Uhlig S. The utility of hotspot mapping for predicting spatial patterns of crime. Security Journal, 2008, 21(1–2): 4-28

[10] Corso, A. J., Alsudais, A., & Hilton, B. (2016). Big social data and GIS: Visualize predictive crime. AMCIS 2016: Surfing the IT Innovation Wave - 22nd Americas Conference on Information Systems, 1–10. https://aisel.aisnet.org/ecis2016_rp/157/

[11] Cressie N, Huang HC. Classes of nonseparable, spatio-temporal stationary covariance functions. Journal of the American Statistical Association, 1999, 94(448): 1330-1339

[12] Da Silva S, Boivin R, Fortin F. Social media as a predictor of urban crime. Criminologie, 2019, 52(2): 83-109

[13] DeVeaux, R. D., Bowman, A. W., & Azzalini, A. (1999). Applied Smoothing Techniques for Data Analysis. In Technometrics (Vol. 41, Issue 3). https://doi.org/10.2307/1270572

[14] Du, Y., & Ding, N. (2023). A Systematic Review of Multi-Scale Spatio-Temporal Crime Prediction Methods. In ISPRS International Journal of Geo-Information (Vol. 12, Issue 6). https://doi.org/10.3390/ijgi12060209

[15] Featherstone, C. (2013). The relevance of social media as it applies in South Africa to crime prediction. In 2013 IST-Africa Conference and Exhibition, IST-Africa 2013 (Issue IST-Africa Conference and Exhibition). https://ieeexplore.ieee.org/abstract/document/6701724

[16] Ferreira, J., Joao, P., & Martins, J. (2012). GIS for Crime Analysis:Geography for Predictive Models. https://www.routledge.com/Spatial-Analysis-and-GIS/Fotheringham-Rogerson/p/book/9780849339337

[17] Gayo-Avello D. A meta-analysis of state-of-the-art electoral prediction from Twitter data. Social Science Computer Review, 2013, 31(6): 649-679

[18] Geoapify. (2024). Geoapify. https://www.geoapify.com/

[19] Gilmour C, Higham DJ. Modelling burglary in Chicago using a self-exciting point process with isotropic triggering. European Journal of Applied Mathematics, 2022, 33(2): 369-391

[20] Goovaerts, P. (1997). Geostatistics for Natural Resources Evaluation. https://global.oup.com/academic/content/series/a/applied-geostatistics-age/?lang=en&cc=us

[21] Hu Y, Wang F, Guin C, Zhu H. A spatio-temporal kernel density estimation framework for predictive crime hotspot mapping and evaluation. Applied Geography, 2018, 99: 89-97

[22] Journel, A. G., & Huijbregts, C. J. (1978). Mining Geostatistics. https://search.worldcat.org/title/800035147

[23] Kadar, C., & Pletikosa, I. (2018). Mining large-scale human mobility data for long-term crime prediction. EPJ Data Science, 7(1). https://doi.org/10.1140/epjds/s13688-018-0150-z

[24] Kyriakidis PC, Journel AG. Geostatistical space-time models: A review. Mathematical Geology, 1999, 31(6): 651-684

[25] Lal S, Tiwari L, Ranjan R, Verma A, Sardana N, Mourya R. Analysis and Classification of Crime Tweets. Procedia Computer Science, 2020, 167(2019): 1911-1919

[26] Lan M, Liu L, Hernandez A, Liu W, Zhou H, Wang Z. The spillover effect of geotagged tweets as a measure of ambient population for theft crime. Sustainability (Switzerland), 2019, 11(23): 1-17

[27] Liu L, Lan M, Eck JE, Yang B, Zhou H. Assessing the intraday variation of the spillover effect of tweets-derived ambient population on crime. Social Science Computer Review, 2022, 40(2): 512-533

[28] Mohamad Zamri NF, Md Tahir N, Megat Ali MSA, Khirul Ashar ND, Al-misreb AA. Mini-review of street crime prediction and classification methods. Jurnal Kejuruteraan, 2021, 33(3): 391-401

[29] Newton, A. D., Hirschfield, A., Armitage, R., Rogerson, M., Monchuk, L., & Wilcox, A. (2008). Evaluation of Licensing Act: Measuring Crime and Disorder in and around Licensed Premises, Research Study SRG/05/007 Annex 2: Birmingham, prepared for the Home Office. July 2007. https://eprints.hud.ac.uk/id/eprint/9546/1/Licensing_Final_Report_March_2008_Supplementary_Annex.pdf?utm_source=chatgpt.com

[30] Okabe A, Satoh T, Sugihara K. A kernel density estimation method for networks, its computational method and a GIS-based tool. International Journal of Geographical Information Science, 2009, 23(1): 7-32

[31] Piña-García CA, Ramírez-Ramírez L. Exploring crime patterns in Mexico City. Journal of Big Data, 2019, 6(1): 65

[32] Rousidis D, Koukaras P, Tjortjis C. Social media prediction: A literature review. Multimedia Tools and Applications, 2020, 79(9–106279-6311

[33] Schoen H, Gayo-Avello D, Takis Metaxas P, Mustafaraj E, Strohmaier M, Gloor P. The power of prediction with social media. Internet Research, 2013, 23(5): 528-543

[34] Shi T, Fu J, Hu X. TSE-tran: Prediction method of telecommunication-network fraud crime based on time series representation and transformer. Journal of Safety Science and Resilience, 2023, 4(4): 340-347

[35] SJPD. (2023a). Crime Statistics - Annual. https://www.sjpd.org/records/crime-stats-maps/crime-statistics-annual

[36] SJPD. (2023b). San Jose Police Department. https://www.sjpd.org/records/documents-policies

[37] Snepvangers JJJC, Heuvelink GBM, Huisman JA. Soil water content interpolation using spatio-temporal kriging with external drift. Geoderma, 2003, 112(3–4): 253-271

[38] Tang, J., Xia, L., & Huang, C. (2023). Explainable Spatio-Temporal Graph Neural Networks. International Conference on Information and Knowledge Management, Proceedings, 2432–2441. https://doi.org/10.1145/3583780.3614871

[39] Tasnim N, Imam IT, Hashem MMA. A novel multi-module approach to predict crime based on multivariate spatio-temporal data using attention and sequential fusion model. IEEE Access, 2022, 10: 48009-48030

[40] U.S. Department of Justice—Federal Bureau of Investigation. (2023). Crime data explorer. NIBRS Estimates. UCR Publications. https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/explorer/crime/nibrs-estimates

[41] Uittenbogaard A, Ceccato V. Space-time clusters of crime in Stockholm, Sweden. Review of European Studies, 2012, 4(5): 148-156

[42] Vomfell L, Härdle WK, Lessmann S. Improving crime count forecasts using Twitter and taxi data. Decision Support Systems, 2018, 113: 73-85

[43] Wang, K., Wang, P., Chen, X., Huang, Q., Mao, Z., & Zhang, Y. (2020). A Feature Generalization Framework for Social Media Popularity Prediction. MM 2020 - Proceedings of the 28th ACM International Conference on Multimedia, 28th ACM International Conference on Multimedia (MM), 4570–4574. https://doi.org/10.1145/3394171.3416294

[44] Wang, Z., Liu, L., Zhou, H., & Lan, M. (2019). Crime geographical displacement: Testing its potential contribution to crime prediction. ISPRS International Journal of Geo-Information, 8(9). https://doi.org/10.3390/ijgi8090383

[45] Yang B, Liu L, Lan M, Wang Z, Zhou H, Yu H. A spatio-temporal method for crime prediction using historical crime data and transitional zones identified from nightlight imagery. International Journal of Geographical Information Science, 2020, 34(9): 1740-1764

[46] Yang D, Heaney T, Tonon A, Wang L, Cudré-Mauroux P. CrimeTelescope: Crime hotspot prediction based on urban and social media data fusion. World Wide Web, 2018, 21(5): 1323-1347

[47] Yu, H., Liu, L., Yang, B., & Lan, M. (2020). Crime Prediction with Historical Crime and Movement Data of Potential Offenders Using a Spatio-Temporal Cokriging Method. ISPRS International Journal of Geo-Information, 9–11. https://www.mdpi.com/2220-9964/9/12/732

[48] Yuan Y, McNeeley S, Melde C. Understanding the fear of crime and perceived risk across immigrant generations: Does the quality of social ties matter?. Crime and Delinquency, 2024, 70(3): 812-843

[49] Yuan Y, Sanchez CV, Punla C. Procedural justice, neighborhood context, and domestic violence reporting intention among subgroups of immigrants. Policing and Society, 2022, 32(10): 1180-1192

[50] Zandiatashbar A, Kayanan CM. Negative consequences of innovation-igniting urban developments: Empirical evidence from three US cities. Urban Planning, 2020, 5(3): 378-391

[51] Zhang, P., Wang, X., & Li, B. (2014). Evaluating Important Factors and Effective Models for Twitter Trend Prediction. In J. Kawash (Ed.), Online Social Media Analysis and Visualization (pp. 81–98). https://doi.org/10.1007/978-3-319-13590-8_5

[52] Zheng X, Han J, Sun A. A survey of location prediction on Twitter. IEEE Transactions on Knowledge and Data Engineering, 2018, 30(9): 1652-1671

Funding

University of New Mexico Office of the Vice President for Research(#8oh6a4x35h)

University of New Mexico Office of the Vice President for Research (#gvvrxwyj64)

University of New Mexico, A&S Interdisciplinary Science Cooperative through the Office of Research (#TA-1003)

RIGHTS & PERMISSIONS

The Author(s)