Explainable AI for Identifying the Environmental Drivers of Public Satisfaction in Urban Spaces
Keywords:
explainable artificial intelligence; urban satisfaction; street-level imagery; spatial confounding; Shapley values; environmental driversAbstract
Urban public satisfaction is shaped by a dense mixture of physical streetscape qualities, accessibility conditions, ambient exposures, and socially produced activity patterns. Machine-learning models can predict satisfaction from such environmental data with considerable flexibility, but predictive accuracy alone does not identify which features should inform design or policy. This manuscript develops a methodological framework for explainable artificial intelligence in urban satisfaction research that treats environmental drivers as spatially embedded, interacting, and uncertainty-qualified quantities rather than isolated ranking scores. The framework fuses street-view imagery, geotagged social media signals, and tabular environmental indicators; estimates satisfaction with a regularized predictive model; explains predictions through Shapley attribution and interaction analysis; diagnoses spatial dependence and scale sensitivity; and converts attributions into uncertainty-aware decision rules. The central contribution is an integrated analytical specification that connects multimodal urban feature construction, model-agnostic explanation, spatial diagnostics, calibration, and intervention screening in one coherent workflow. The framework also clarifies the difference between predictive attribution and causal effect, explains why tree-based models and conditional attribution are often preferable for tabular environmental features, and identifies failure modes associated with correlated variables, spatial confounding, platform bias, and the modifiable areal unit problem. Practical significance lies in producing driver rankings and interaction profiles that can be audited, spatially inspected, and translated into design priorities without overstating what a predictive model can establish. The approach supports more defensible use of explainable machine learning in urban space assessment and public satisfaction research.
References
1. Marans, R. W., & Stimson, R. J. (2011). Investigating quality of urban life: Theory, methods, and empirical research. Springer.
2. Ewing, R., & Handy, S. (2009). Measuring the unmeasurable: Urban design qualities related to walkability. Journal of Urban Design, 14(1), 65–84.
3. Rundle, A. G., Bader, M. D. M., Richards, C. A., Neckerman, K. M., & Teitler, J. O. (2011). Using Google Street View to audit neighborhood environments. American Journal of Preventive Medicine, 40(1), 94–100.
4. Zheng, Y., Liu, T., Wang, Y., Zhu, Y., Liu, Y., & Chang, E. (2014). Diagnosing New York City's noises with ubiquitous data. In Proceedings of the 2014 ACM International Joint Conference on Pervasive and Ubiquitous Computing (pp. 715–725). ACM.
5. Aman, J., Matisziw, T. C., Kim, J. B., & Luo, D. (2022, April). Sensing the city: Leveraging geotagged social media posts and street view imagery to model urban streetscapes using deep neural networks. In Proceedings of the International Conference on Computer-Aided Architectural Design Research in Asia (pp. 595-604). Association for Computer-Aided Architectural Design Research in Asia.
6. Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. In Advances in Neural Information Processing Systems 30 (pp. 4765–4774). Curran Associates.
7. Ribeiro, M. T., Singh, S., & Guestrin, C. (2016). "Why should I trust you?" Explaining the predictions of any classifier. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (pp. 1135–1144). ACM.
8. Doshi-Velez, F., & Kim, B. (2017). Towards a rigorous science of interpretable machine learning (arXiv:1702.08608). arXiv.
9. Molnar, C. (2022). Interpretable machine learning: A guide for making black box models explainable (2nd ed.). Independently published.
10. LeCun, Y., Bengio, Y., & Hinton, G. (2015). Deep learning. Nature, 521(7553), 436–444.
11. Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT Press.
12. Breiman, L. (2001). Random forests. Machine Learning, 45(1), 5–32.
13. Grinsztajn, L., Oyallon, E., & Varoquaux, G. (2022). Why do tree-based models still outperform deep learning on typical tabular data? In Advances in Neural Information Processing Systems 35. Curran Associates.
14. Anselin, L. (1995). Local indicators of spatial association—LISA. Geographical Analysis, 27(2), 93–115.
15. Fotheringham, A. S., Brunsdon, C., & Charlton, M. (2002). Geographically weighted regression: The analysis of spatially varying relationships. Wiley.
16. Openshaw, S. (1984). The modifiable areal unit problem. Geo Books.
17. Steegen, S., Tuerlinckx, F., Gelman, A., & Vanpaemel, W. (2016). Increasing transparency through a multiverse analysis. Perspectives on Psychological Science, 11(5), 702–712.
18. Hastie, T., Tibshirani, R., & Friedman, J. (2009). The elements of statistical learning: Data mining, inference, and prediction (2nd ed.). Springer.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Computational Intelligence Systems

This work is licensed under a Creative Commons Attribution 4.0 International License.
This article is published under the Creative Commons Attribution 4.0 International License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.