Spatial-Temporal Mixture-of-Graph-Experts for Multi-Type Crime Prediction

As various types of crimes continue to threaten public safety and economic development, predicting the occurrence of multiple types of crimes becomes increasingly vital for effective prevention measures. Although extensive efforts have been made, most of them overlook the heterogeneity of different crime categories and fail to address the issue of imbalanced spatial distribution. In this work, we propose Spatial-Temporal Mixture-of-Graph-Experts (ST-MoGE), a framework for collective multiple-type crime prediction. To enhance the model's ability to identify diverse spatial-temporal dependencies and mitigate potential conflicts caused by spatial-temporal heterogeneity of different crime categories, we introduce a module, Mixture-of-Graph-Experts (MGE), to capture the distinctive and shared crime patterns of each crime category. Then, we propose Cross-Expert Contrastive Learning (CECL) to refine MGE and force each expert to specialize in modeling specific patterns, thereby reducing blending and redundancy. Furthermore, to address the issue of imbalanced spatial distribution, we propose a module, Hierarchical Adaptive Loss Re-Weighting (HALR), to eliminate biases and underfitting in data-scarce regions. To evaluate the effectiveness of our methods, we conduct comprehensive experiments on two real-world crime datasets and compare our results with 12 advanced baselines. The experimental results demonstrate the superiority of our methods.