A graph-based framework for analysing human mobility using spatial-temporal network representations

doi:10.1080/14498596.2026.2664052

A graph-based framework for analysing human mobility using spatial-temporal network representations

2026-04-03

0

PRE

AI

DOI：10.1080/14498596.2026.2664052

Original

Original request for help

Save

Abstract

En 中文

Predicting human mobility requires choices about how to represent people, places, and time. This paper provides a graph-construction and task-formulation framework for mobility analysis using passively collected trajectories. Using the Humob 2023 dataset, we instantiate three commonly used graph representations then use Graph Attention Networks (GAT) and Graph Convolutional Network (GCN) baselines to separate representation effects from architectural effects. The results indicate that representation and task definition dominate performance differences, while the gap between GAT and GCN is comparatively smaller in this setting. We distil practical considerations for selecting mobility graph representations and discuss limitations, especially the need for explicit dynamic-graph or sequence modelling when fine-grained temporal evolution is central.

Keywords:

Human mobility

graph neural networks

graph attention networks

geospatial AI

A graph-based framework for analysing human mobility using spatial-temporal network representations

Abstract

Journal

Organization