Miaoxuan Shan

Accurate crime spatiotemporal prediction is crucial for crime prevention. However, crime occurrences are influenced by diverse and interacting social factors, resulting in dynamically evolving distributions with non-stationarity and spatial heterogeneity. Most existing methods focus on data preprocessing or architectural enhancements and remain offline models, which limits their generalization capability. To address these challenges, we propose a novel spatial semantic-guided online learning framework. Specifically, we first compute the spatial semantic similarity between urban regions using points of interest. Based on this, we then introduce a contrastive learning objective guided by this similarity during training. This design aims to enhance the model’s ability to capture both the similarities and discrepancies among regions. During the prediction process, an iterative online learning strategy is employed to adapt to dynamically changing crime patterns. By continuously fine-tuning the model with streaming data, the proposed framework improves robustness and generalization under non-stationary crime spatiotemporal distributions. Finally, extensive experiments on real-world crime datasets indicate the effectiveness and stability of our proposed approach.

Objective crime risk and perceived safety constitute distinct yet interrelated dimensions of urban security, whose spatial discrepancies may lead to misaligned policy interventions. This study develops a street-level analytical framework to examine the (mis)match between perceived safety and crime risk in Chaoyang District, Beijing. An enhanced Street-view imagery (SVI) segmentation model with object detection was applied to extract streetscape elements and estimate perceived safety scores, which were then standardized and compared with street-level crime data, yielding two types of matches and two types of mismatches. Three conditions were analyzed using multinomial logit regression: (1) objective unsafety with low perceived safety, (2) objective safety with low perceived safety, and (3) objective unsafety with high perceived safety. Findings demonstrate how visual and social environmental factors jointly shape discrepancies between perceived and actual safety and identify potential determinants to mitigate such (mis)matches.

Current crime spatiotemporal prediction models are limited by the insufficient ability of POI data to represent the continuity and mixed-use nature of urban spatial functions. To address this, our study applies an urban region representation method based on building footprints and validates its effectiveness in improving the accuracy of crime spatiotemporal prediction. Specially, we first use the Region Dual Contrastive Learning algorithm to generate region representations as a region graph by integrating building footprints and POI data. Then, the region graph combined with crime data is input into crime prediction models to predict four crime types, including Burglary, Robbery, Felony Assault, and Grand Larceny. Finally, ablation experiments are conducted to quantify the contribution of building footprints to prediction improvement. The experimental results on New York City crime data indicate that (1) the region representations significantly improve deep learning model performance, with the most improved LSTM achieving average increases of 5.66% in Macro-F1 and 18.57% in Micro-F1, particularly benefiting baseline models with lower accuracy, and (2) the region representations yield more significant improvements for low-frequency crime categories and mitigates temporal memory decay in long-term predictions. These findings confirm that incorporating urban region representation based on building footprints effectively enhances crime spatiotemporal prediction performance, providing a more precise and efficient tool for urban security management to optimize police resource allocation and crime prevention strategies.