Using publicly available data for predicting socioeconomic values in urban context

Abstract

Urban transportation networks are recognized for their pivotal role in forecasting city indicators and facilitating efficient planning and management. However, despite the increase of methodologies and models harnessing machine learning advancement to forecast these values, challenges persist in scenarios where direct demographic or economic data are limited or unavailable. In this paper, we propose an approach to infer socioeconomic information in an urban context without relying on traditional, official data sources, but rather focusing on publicly available data relating to the digital footprints of the cities’ inhabitants. We leverage Graph Neural Network (GNN) models to capture the spatial relationships inherent in network data while integrating perceptual features extracted from images to enhance predictive accuracy. Our results demonstrate that the combination of these data sources enables a GNN to achieve robust performance in predicting socioeconomic indicators, particularly in settings where traditional demographic and economic data may be sparse or unavailable. Through our analysis, we show that while perceptual features alone offer substantial predictive power, the inclusion of map topology through GNN models provides crucial context, leading to better generalization and more reliable predictions across different urban areas.

Keywords

Graph neural networks / Urban transportation / Perceptual values / Socioeconomic forecasting

References

[1] Asadi BagloeeS, HeshmatiM, TavanaM, Di CaprioD. A logit-based model for measuring the effects of transportation infrastructure on land value. Transportation Planning and Technology, 2017, 40(2): 143-166

[2] ChaturvediV, de VriesWT. Machine learning algorithms for urban land use planning: A review. Urban Science, 2021, 5(3): 68

[3] Chen, L., Xu, J., Wu, B., Qian, Y., Du, Z., Li, Y., & Zhang, Y. (2021). Group-aware graph neural network for nationwide city air quality forecasting (2021). arXiv preprint arXiv:2108.12238.

[4] Dubey, A., Naik, N., Parikh, D., Raskar, R., & Hidalgo, C. A. (2016). Deep learning the city: Quantifying urban perception at a global scale. In Computer Vision–ECCV 2016: 14th European Conference, Amsterdam, The Netherlands, October 11–14, 2016, Proceedings, Part I 14 (pp. 196–212). Springer.

[5] Hamilton, W., Ying, Z., & Leskovec, J. (2017). Inductive representation learning on large graphs. Advances in Neural Information processing Systems, 30, 1024-1034.

[6] Hochreiter, S. (1997). Long short-term memory. Neural Computation MIT-Press.

[7] HurtubiaR, GuevaraA, DonosoP. Using images to measure qualitative attributes of public spaces through sp surveys. Transportation Research Procedia, 2015, 11: 460-474

[8] IbeasÁ, CorderaR, Dell’OlioL, CoppolaP, DominguezA. Modelling transport and real-estate values interactions in urban systems. Journal of Transport Geography, 2012, 24: 370-382

[9] Iglesias, P., Greene, M., & Ortúzar, J. de D. (2013). On the perception of safety in low income neighbourhoods: using digital images in a stated choice experiment. In S. Hess, & A. J. Daly (Eds.), Choice Modelling: The State of the Art and the State of Practice (pp. 193–210). Edward Elgar Publishing Ltd.

[10] Kipf, T. N., & Welling, M. (2016). Semi-supervised classification with graph convolutional networks. arXiv preprint arXiv:1609.02907.

[11] Kramer, O., spsampsps Kramer, O. (2016). Scikit-learn. Machine Learning for Evolution Strategies, 20 45–53.

[12] LaingR, DaviesA-M, MillerD, ConniffA, ScottS, MorriceJ. The application of visual environmental economics in the study of public preference and urban greenspace. Environment and Planning B: Planning and Design, 2009, 36(2): 355-375

[13] Lewis, M. (2019). Bart: Denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension. arXiv preprint arXiv:1910.13461.

[14] Munasinghe, T., & Behlendorf, B. (2022). Using graph neural networks to investigate the relationship between the socioeconomic factors and emergency medical service (ems) median response time in new york city. In 2022 IEEE International Conference on Big Data (Big Data) (pp. 6781–6783). IEEE.

[15] NasarJL. The evaluative image of the city. Journal of the American Planning Association, 1990, 56(1): 41-53

[16] NavarroI, GalileaP, HidalgoR, HurtubiaR. Transporte y su integración con el entorno urbano: ¿ cómo incorporamos los beneficios de elementos urbanos en la evaluación de proyectos de transporte?. EURE (Santiago), 2018, 44(132): 135-153

[17] RaffelC, ShazeerN, RobertsA, LeeK, NarangS, MatenaM, ZhouY, LiW, LiuPJ. Exploring the limits of transfer learning with a unified text-to-text transformer. Journal of Machine Learning Research, 2020, 21(140): 1-67

[18] RossettiT, LobelH, RoccoV, HurtubiaR. Explaining subjective perceptions of public spaces as a function of the built environment: A massive data approach. Landscape and Urban Planning, 2019, 181: 169-178

[19] Rossi, E., Kenlay, H., Gorinova, M. I., Chamberlain, B. P., Dong, X., & Bronstein, M. M. (2022). On the unreasonable effectiveness of feature propagation in learning on graphs with missing node features. In Learning on graphs conference (pp. 11–1). PMLR.

[20] SalessesP, SchechtnerK, HidalgoCA. The collaborative image of the city: Mapping the inequality of urban perception. PloS One, 2013, 8(7): e68400

[21] UN-Habitat (2024). World smart cities outlook 2024. https://unhabitat.org/world-smart-cities-outlook-2024. Accessed 17 Apr 2025.

[22] Vaswani, A. (2017). Attention is all you need. Advances in Neural Information Processing Systems, 30, 5998-6008.

[23] Veličković, P., Cucurull, G., Casanova, A., Romero, A., Lio, P., & Bengio, Y. (2017). Graph attention networks. arXiv preprint arXiv:1710.10903.

[24] World Bank (2021). World development report 2021: Data for better lives. https://www.worldbank.org/en/publication/wdr2021. Accessed 17 Apr 2025.

[25] Zhang, J., Shi, X., Xie, J., Ma, H., King, I., & Yeung, D.-Y. (2018). Gaan: Gated attention networks for learning on large and spatiotemporal graphs. arXiv preprint arXiv:1803.07294.

[26] Zhang, W., Liu, H., Liu, Y., Zhou, J., & Xiong, H. (2020). Semi-supervised hierarchical recurrent graph neural network for city-wide parking availability prediction. In Proceedings of the AAAI Conference on Artificial Intelligence (vol. 34, pp. 1186–1193). Association for the Advancement of Artificial Intelligence.

[27] ZhangY, XieZ, JiangF, XuT, YangS, YinS, ZhaoF, HeJ, ChenY, HouZ, HaiyunJ. Evaluating the socioeconomic value of urban underground space in kunming, china, using the entropy method and exponential smoothing prediction. Journal of Urban Planning and Development, 2023, 149(2): 05023001

Funding

Agencia Nacional de Investigación y Desarrollo(ICN17_002)

Fondo Nacional de Desarrollo Científico y Tecnológico(1221799)

RIGHTS & PERMISSIONS

The Author(s)