No abstract

The crime forecasting is an important problem as it greatly contributes to urban safety. Typically, the goal of the problem is to predict different types of crimes for each geographical region (like a neighborhood or censor tract) in the near future. Since nearby regions usually have similar socioeconomic characteristics which indicate similar crime patterns, recent state-of-the-art solutions constructed a distance-based region graph and utilized Graph Neural Network (GNN) techniques for crime forecasting, because the GNN techniques could effectively exploit the latent relationships between neighboring region nodes in the graph if the edges reveal high dependency or correlation. However, this distance-based predefined graph cannot fully capture crime correlation between regions that are far from each other but share similar crime patterns. Hence, to make an accurate crime prediction, the main challenge is to learn a better graph that reveals the dependencies between regions in crime occurrences and meanwhile captures the temporal patterns from historical crime records. To address these challenges, we propose an endto-end graph convolutional recurrent network called HAGEN with several novel designs for crime prediction. Specifically, our framework could jointly capture the crime correlation between regions and the temporal crime dynamics by combining an adaptive region graph learning module with the Diffusion Convolution Gated Recurrent Unit (DCGRU). Based on the homophily assumption of GNN (i.e., graph convolution works better where neighboring nodes share the same label), we propose a homophily-aware constraint to regularize the optimization of the region graph so that neighboring region nodes on the learned graph share similar crime patterns, thus fitting the mechanism of diffusion convolution. It also incorporates crime embedding to model the interdependencies between regions and crime categories. Empirical experiments and comprehensive analysis on two real-world datasets showcase the effectiveness of HAGEN.Preprint. Under review.

Can health conditions be inferred from an individual's mobility pattern? Existing research has discussed the relationship between individual physical activity/mobility and well-being, yet no systematic study has been done to investigate the predictability of fine-grained health conditions from mobility, largely due to the unavailability of data and unsatisfactory modelling techniques. Here, we present a large-scale longitudinal study, where we collect the health conditions of 747 individuals who visit a hospital and tracked their mobility for 2 months in Beijing, China. To facilitate fine-grained individual health condition sensing, we propose HealthWalks, an interpretable machine learning model that takes user location traces, the associated points of interest, and user social demographics as input, at the core of which a Deterministic Finite Automaton (DFA) model is proposed to auto-generate explainable features to capture useful signals. We evaluate the effectiveness of our proposed model, which achieves 40.29% in micro-F1 and 31.63% in Macro-F1 for the 8-class disease category prediction, and outperforms the best baseline by 22.84% in Micro-F1 and 31.79% in Macro-F1. In addition, deeper analysis based on the SHapley Additive exPlanations (SHAP) showcases that HealthWalks can derive meaningful insights with regard to the correlation between mobility and health conditions, which provide important research insights and design implications for mobile sensing and health informatics.

The goal of the crime forecasting problem is to predict different types of crimes for each geographical region (like a neighborhood or censor tract) in the near future. Since nearby regions usually have similar socioeconomic characteristics which indicate similar crime patterns, recent state-of-the-art solutions constructed a distance-based region graph and utilized Graph Neural Network (GNN) techniques for crime forecasting, because the GNN techniques could effectively exploit the latent relationships between neighboring region nodes in the graph if the edges reveal high dependency or correlation. However, this distance-based pre-defined graph can not fully capture crime correlation between regions that are far from each other but share similar crime patterns. Hence, to make a more accurate crime prediction, the main challenge is to learn a better graph that reveals the dependencies between regions in crime occurrences and meanwhile captures the temporal patterns from historical crime records. To address these challenges, we propose an end-to-end graph convolutional recurrent network called HAGEN with several novel designs for crime prediction. Specifically, our framework could jointly capture the crime correlation between regions and the temporal crime dynamics by combining an adaptive region graph learning module with the Diffusion Convolution Gated Recurrent Unit (DCGRU). Based on the homophily assumption of GNN (i.e., graph convolution works better where neighboring nodes share the same label), we propose a homophily-aware constraint to regularize the optimization of the region graph so that neighboring region nodes on the learned graph share similar crime patterns, thus fitting the mechanism of diffusion convolution. Empirical experiments and comprehensive analysis on two real-world datasets showcase the effectiveness of HAGEN.

No abstract