Hybrid Spatio-Temporal Crime Forecasting with Graph Neural Networks

Orekoya; V P; Mathias D; Bennett; EO; Anireh; VIE

doi:10.56201/ijcsmt.vol.11.no9.2025.pg132.143

📄 Download Paper

Hybrid Spatio-Temporal Crime Forecasting with Graph Neural Networks

Orekoya, V P, Mathias D, Bennett, EO, Anireh, VIE

Abstract

Traditional crime forecasting models often struggle with the complex, interconnected dynamics of urban crime, primarily because they analyze spatial and temporal factors in isolation. This paper introduces a novel hybrid deep learning framework designed to overcome these limitations by modeling crime as a dynamic spatio-temporal neural network. Our approach integrates a Spatio- Temporal Graph Neural Network (ST-GNN) to capture complex inter-location dependencies and a robust Enhanced Long Short-Term Memory (eLSTM) to model intricate temporal patterns. This synergistic fusion enables a more holistic understanding of crime propagation and frequency. We evaluate this model on a five-year urban crime dataset, where it demonstrates superior performance over standalone forecasting methods, achieving a 10–15% reduction in Mean Absolute Error (MAE) and a hotspot precision of 0.88. The results of our ablation studies underscore the importance of combining these complementary learning paradigms. The proposed framework provides a highly accurate and scalable tool for proactive policing and resource allocation.

Keywords

Crime ForecastingSpatio-temporal AnalysisGraph Neural Networks (GNNs)

References

[1] Chainey, S., et al. (2002). GIS-Based Crime Hotspot Analysis: A Comparative Study of Techniques. In Proceedings of the 5th Annual International Conference on Geographic Information Science. [2] Scarselli, F., et al. (2009). The graph neural network model. IEEE Transactions on Neural Networks, 20(1), 61-80. [3] Yu, B., et al. (2018). Spatio-temporal graph convolutional networks: A deep learning framework for traffic forecasting. In Proceedings of the 27th International Joint Conference on Artificial Intelligence (IJCAI). [4] Liu, Y., et al. (2019). Crime prediction with an enhanced LSTM network and attention mechanism. Journal of Quantitative Criminology, 35(3), 481-502. [5] Geng, X., et al. (2019). Spatio-temporal graph convolutional networks for ride-hailing demand forecasting. In Proceedings of the AAAI Conference on Artificial Intelligence. [6] Geng, X., et al. (2019). Spatio-temporal graph convolutional networks for ride-hailing demand forecasting. In Proceedings of the AAAI Conference on Artificial Intelligence. [7] Geng, X., et al. (2018). A spatio-temporal deep learning approach for traffic flow forecasting. In Proceedings of the 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. [8] Wang, X., & Brown, D. E. (2011). The spatio-temporal generalized additive model for criminal incidents. In Proceedings of the 2011 IEEE International Conference on Intelligence and Security Informatics. [9] Wang, J., et al. (2019). Crime prediction with spatio-temporal convolutional neural networks. Journal of Urban Analytics, 2(1), 1-15. [10] Grubesic, T. H. (2006). A comparison of density-based and point-based methods for identifying crime hot spots. Journal of Geographic Information Science, 10(2), 173-195.

Hybrid Spatio-Temporal Crime Forecasting with Graph Neural Networks

Abstract

Keywords

References

More Articles from INTERNATIONAL JOURNAL OF COMPUTER SCIENCE AND MATHEMATICAL THEORY