Commute Networks as a Signature of Urban Socioeconomic Performance: Evaluating Mobility Structures with Deep Learning Models

2025-07-05

AI Summary

Conventional urban socioeconomic modeling relies heavily on geographic proximity and node-level attributes (e.g., population, GDP), overlooking the network effects embodied in inter-city commuting patterns. Method: We propose a purely structure-driven modeling paradigm that constructs an inter-city mobility network solely from census-based commuting flow data; the network’s topological structure serves as a “structural signature,” with no node attributes incorporated. We design an end-to-end supervised framework integrating a graph neural network (GNN) and a feedforward network to directly predict city-level socioeconomic indicators. Contribution/Results: Experiments across 12 U.S. metropolitan areas demonstrate that our approach significantly outperforms conventional machine learning baselines. This work is the first to empirically validate—within a unified deep learning framework—that the structural topology of commuting networks alone carries strong predictive power for urban socioeconomic outcomes, establishing a novel attribute-free paradigm for urban modeling.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Social Networks and Social Media: Computational social science
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

Urban socioeconomic modeling has predominantly concentrated on extensive location and neighborhood-based features, relying on the localized population footprint. However, networks in urban systems are common, and many urban modeling methods don't account for network-based effects. In this study, we propose using commute information records from the census as a reliable and comprehensive source to construct mobility networks across cities. Leveraging deep learning architectures, we employ these commute networks across U.S. metro areas for socioeconomic modeling. We show that mobility network structures provide significant predictive performance without considering any node features. Consequently, we use mobility networks to present a supervised learning framework to model a city's socioeconomic indicator directly, combining Graph Neural Network and Vanilla Neural Network models to learn all parameters in a single learning pipeline. Our experiments in 12 major U.S. cities show the proposed model outperforms previous conventional machine learning models. This work provides urban researchers methods to incorporate network effects in urban modeling and informs stakeholders of wider network-based effects in urban policymaking and planning.

Problem

• Model urban socioeconomic performance using commute networks
• Incorporate network effects in urban modeling with deep learning
• Predict socioeconomic indicators without relying on node features

Innovation

• Using commute networks for urban socioeconomic modeling
• Combining Graph and Vanilla Neural Networks
• Predicting performance without node features