From Points to Places: Towards Human Mobility-Driven Spatiotemporal Foundation Models via Understanding Places

2025-06-17

AI Summary

Human mobility data exhibits high complexity across spatial, temporal, and semantic dimensions, yet existing models predominantly rely on point-based POIs, failing to capture semantically rich, behaviorally grounded dynamic places. Method: We propose the first spatiotemporal foundation model centered on *places*—not POIs—as the fundamental modeling unit. Our approach integrates geographic-semantic knowledge, multi-granularity spatiotemporal dynamics modeling, and human-behavior-driven mechanisms within a graph neural network–based self-supervised pretraining framework, enabling hierarchical place representation learning and context-aware cross-domain transfer. Contribution/Results: The model overcomes limitations of point-wise modeling by unifying the understanding of dynamic, semantically meaningful, and behaviorally interpretable places. Experiments demonstrate substantial improvements in cross-regional generalization and decision-making efficiency across tasks including personalized place discovery, urban planning, and logistics optimization—establishing a new paradigm for scalable, transferable geospatial intelligence.

Technology Category

• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Capturing human mobility is essential for modeling how people interact with and move through physical spaces, reflecting social behavior, access to resources, and dynamic spatial patterns. To support scalable and transferable analysis across diverse geographies and contexts, there is a need for a generalizable foundation model for spatiotemporal data. While foundation models have transformed language and vision, they remain limited in handling the unique challenges posed by the spatial, temporal, and semantic complexity of mobility data. This vision paper advocates for a new class of spatial foundation models that integrate geolocation semantics with human mobility across multiple scales. Central to our vision is a shift from modeling discrete points of interest to understanding places: dynamic, context-rich regions shaped by human behavior and mobility that may comprise many places of interest. We identify key gaps in adaptability, scalability, and multi-granular reasoning, and propose research directions focused on modeling places and enabling efficient learning. Our goal is to guide the development of scalable, context-aware models for next-generation geospatial intelligence. These models unlock powerful applications ranging from personalized place discovery and logistics optimization to urban planning, ultimately enabling smarter and more responsive spatial decision-making.

Problem

• Develop a generalizable foundation model for spatiotemporal mobility data
• Address spatial, temporal, and semantic complexity in mobility modeling
• Shift from discrete points to dynamic, context-rich places modeling

Innovation

• Spatial foundation models integrating geolocation semantics
• Shift from discrete points to dynamic places modeling
• Scalable context-aware models for geospatial intelligence