GeoPro-Net: Learning Interpretable Spatiotemporal Prediction Models through Statistically-Guided Geo-Prototyping
Bang An, Xun Zhou, Zirui Zhou, Ronilo Ragodos, Zenglin Xu, Jun Luo
TL;DR
GeoPro-Net tackles the interpretability gap in spatiotemporal event forecasting by introducing statistically guided Geo-concepts and a prototype-based reasoning framework. It extracts multi-scale, interpretable patterns via SSCE, consolidates them with geo-aware pooling, and grounds predictions in learned prototypes that can be projected onto real-world cases. The approach achieves faithful interpretability with competitive accuracy on Chicago and NYC crime and accident datasets, outperforming several black-box baselines and offering intuitive explanations through prototype mappings and regional concept scores. This work demonstrates that intrinsically interpretable spatiotemporal models can support urban decision-making with transparent, case-based reasoning.
Abstract
The problem of forecasting spatiotemporal events such as crimes and accidents is crucial to public safety and city management. Besides accuracy, interpretability is also a key requirement for spatiotemporal forecasting models to justify the decisions. Interpretation of the spatiotemporal forecasting mechanism is, however, challenging due to the complexity of multi-source spatiotemporal features, the non-intuitive nature of spatiotemporal patterns for non-expert users, and the presence of spatial heterogeneity in the data. Currently, no existing deep learning model intrinsically interprets the complex predictive process learned from multi-source spatiotemporal features. To bridge the gap, we propose GeoPro-Net, an intrinsically interpretable spatiotemporal model for spatiotemporal event forecasting problems. GeoPro-Net introduces a novel Geo-concept convolution operation, which employs statistical tests to extract predictive patterns in the input as Geo-concepts, and condenses the Geo-concept-encoded input through interpretable channel fusion and geographic-based pooling. In addition, GeoPro-Net learns different sets of prototypes of concepts inherently, and projects them to real-world cases for interpretation. Comprehensive experiments and case studies on four real-world datasets demonstrate that GeoPro-Net provides better interpretability while still achieving competitive prediction performance compared with state-of-the-art baselines.