A spatio-temporal method for crime prediction using historical crime data and transitional zones identified from nightlight imagery

ABSTRACT

Accurate crime prediction can help allocate police resources for crime reduction and prevention. There are two popular approaches to predict criminal activities: one is based on historical crime, and the other is based on environmental variables correlated with criminal patterns. Previous research on geo-statistical modeling mainly considered one type of data in space-time domain, and few sought to blend multi-source data. In this research, we proposed a spatio-temporal Cokriging algorithm to integrate historical crime data and urban transitional zones for more accurate crime prediction. Time-series historical crime data were used as the primary variable, while urban transitional zones identified from the VIIRS nightlight imagery were used as the secondary co-variable. The algorithm has been applied to predict weekly-based street crime and hotspots in Cincinnati, Ohio. Statistical tests and Predictive Accuracy Index (PAI) and Predictive Efficiency Index (PEI) tests were used to validate predictions in comparison with those of the control group without using the co-variable. The validation results demonstrate that the proposed algorithm with historical crime data and urban transitional zones increased the correlation coefficient by 5.4% for weekdays and by 12.3% for weekends in statistical tests, and gained higher hit rates measured by PAI/PEI in the hotspots test.

KEYWORDS:

• Crime prediction
• spatio-temporal modeling
• Cokriging
• VIIRS nightlight

Acknowledgments

The authors would like to acknowledge the Cincinnati policy department for the provision of the crime data. They are also grateful to NASA Earthdata for distributing and processing the NPP-VIIRS nightlight imagery.

Disclosure statement

No potential conflict of interest was reported by the authors.

Data and codes availability statement

The data and codes that support the findings of this study are available in Mendeley data with the Reserved DOI: doi:10.17632/gnbk75nzxm.1 Codes are available on the Github crime-prediction repository. Crime points data with address in the folder of ‘Points_Data_Private’ cannot be made publicly available to protect research participant privacy and consent.

Additional information

Funding

This work was supported by the University of Central Florida Preeminent Postdoctoral Program.

Notes on contributors

Bo Yang

Dr. Bo Yang is an interdisciplinary postdoctoral researcher of GIScience at the University of Central Florida. His research interests are: GIScience, spatial statistics, machine learning algorithms, environmental and sociological modeling, UAV & drone coastal mapping.

Lin Liu

Dr. Lin Liu is a professor of geography.  His main research interests include crime analysis, urban informatics, GIS and remote sensing. He has published over 200 journal articles.

Minxuan Lan

Minxuan Lan is a Ph.D. candidate at Department of Geography and GIS at the University of Cincinnati. His main research interests include crime analysis, GIS, big data and public health.

Zengli Wang

Dr. Zengli Wang is an associate professor at Nanjing Forestry University, China. His research focuses on crime prediction, crime pattern analysis and spatial data mining.

Hanlin Zhou

Hanlin Zhou is a master student in geography at the University of Cincinnati.  His main research interests include crime research and urban informatics.

Hongjie Yu

Hongjie YU is a PhD student at Sun Yat-sen University, Guangzhou, China. Her main research areas include crime geography and crime analysis.