Explainable Deep Learning Framework for Predicting Urban Air Quality from Multisource Data Streams
Keywords:
air quality prediction, explainable deep learning, multisource data streams, urban informatics, governance, fairness, spatiotemporal modelingAbstract
Urban air quality prediction has emerged as a critical analytics problem demanding integration of heterogeneous data streams, high-fidelity modeling, and actionable explanations for decision support. This paper presents a comprehensive framework for an explainable deep learning system that ingests continuous multisource data, including regulatory monitoring stations, low-cost sensor networks, meteorological feeds, traffic telemetry, and satellite-derived aerosol optical depth, to produce high-resolution air quality forecasts. The architecture couples a spatiotemporal attention-based deep learning predictor with a suite of post-hoc and intrinsic interpretability modules, enabling the provision of human-readable rationales alongside each forecast. We discuss structural trade-offs in designing a system that balances prediction latency, model capacity, and faithfulness of explanations under resource constraints typical of municipal deployments. The governance architecture embedding the framework is examined with regard to data provenance, model versioning, and auditability, while robustness is analyzed under sensor drift, data missingness, and adversarial perturbation scenarios. Fairness considerations are foregrounded, highlighting how differential sensor density and model opacity can produce disparate impacts on marginalized communities. Policy implications for integrating explainable forecasts into public health alerting, traffic management, and urban planning are evaluated. The paper argues that high-accuracy prediction alone is insufficient; sustainable sociotechnical infrastructure requires built-in transparency mechanisms that align model reasoning with regulatory standards and community trust. A longitudinal deployment perspective is maintained, addressing operational sustainability, calibration decay, and institutional capacity-building.
References
1. Huang, C.-J., & Kuo, P.-H. (2018). A deep CNN-LSTM model for particulate matter (PM2.5) forecasting in smart cities. Sensors, 18(7), 2220.
2. Li, X., Peng, L., Yao, X., Cui, S., Hu, Y., You, C., & Chi, T. (2017). Long short-term memory neural network for air pollutant concentration predictions: Method development and evaluation. Environmental Pollution, 231, 997–1004.
3. Wang, J., & Song, G. (2018). A deep spatial-temporal ensemble model for air quality prediction. Neurocomputing, 314, 198–206.
4. Zhao, J., Deng, F., Cai, Y., & Chen, J. (2019). Long short-term memory-Fully connected (LSTM-FC) neural network for PM2.5 concentration prediction. Chemosphere, 220, 486–492.
5. Liang, Y., Ke, S., Zhang, J., Yi, X., & Zheng, Y. (2018). GeoMAN: Multi-level attention networks for geo-sensory time series prediction. Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence, 3428–3434.
6. Zheng, Y., Yi, X., Li, M., Li, R., Shan, Z., Chang, E., & Li, T. (2015). Forecasting fine-grained air quality based on big data. Proceedings of the 21st ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 2267–2276.
7. Sundararajan, M., Taly, A., & Yan, Q. (2017). Axiomatic attribution for deep networks. Proceedings of the 34th International Conference on Machine Learning, 3319–3328.
8. Montavon, G., Binder, A., Lapuschkin, S., Samek, W., & Müller, K.-R. (2019). Layer-wise relevance propagation: An overview. In Explainable AI: Interpreting, Explaining and Visualizing Deep Learning (pp. 193–209). Springer.
9. Ribeiro, M. T., Singh, S., & Guestrin, C. (2016). “Why should I trust you?” Explaining the predictions of any classifier. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 1135–1144.
10. Lundberg, S. M., & Lee, S.-I. (2017). A unified approach to interpreting model predictions. Advances in Neural Information Processing Systems, 30, 4765–4774.
11. Jain, S., & Wallace, B. C. (2019). Attention is not explanation. Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 3543–3556.
12. Arrieta, A. B., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., ... & Herrera, F. (2020). Explainable artificial intelligence (XAI): Concepts, taxonomies, opportunities and challenges toward responsible AI. Information Fusion, 58, 82–115.
13. Floridi, L., Cowls, J., Beltrametti, M., Chatila, R., Chazerand, P., Dignum, V., ... & Schafer, B. (2018). AI4People—An ethical framework for a good AI society: Opportunities, risks, principles, and recommendations. Minds and Machines, 28(4), 689–707.
14. Mitchell, M., Wu, S., Zaldivar, A., Barnes, P., Vasserman, L., Hutchinson, B., ... & Gebru, T. (2019). Model cards for model reporting. Proceedings of the Conference on Fairness, Accountability, and Transparency, 220–229.
15. Veeramachaneni, K., Vladen, I., & Re, C. (2016). Deep feature synthesis: Towards automating data science endeavors. Proceedings of the IEEE International Conference on Data Science and Advanced Analytics, 182–191.
16. Ma, J., Ding, Y., Cheng, J. C. P., Jiang, F., & Xu, Z. (2019). Soft detection of anomalous sensor data in IoT-based air quality monitoring systems. IEEE Internet of Things Journal, 6(6), 10776–10786.
17. Doshi-Velez, F., & Kim, B. (2017). Towards a rigorous science of interpretable machine learning. arXiv preprint arXiv:1702.08608.
18. Selbst, A. D., Boyd, D., Friedler, S. A., Venkatasubramanian, S., & Vertesi, J. (2019). Fairness and abstraction in sociotechnical systems. Proceedings of the Conference on Fairness, Accountability, and Transparency, 59–68.
19. Zou, J., & Schiebinger, L. (2018). AI can be sexist and racist—it’s time to make it fair. Nature, 559, 324–326.
20. Holzinger, A., Langs, G., Denk, H., Zatloukal, K., & Müller, H. (2019). Causability and explainability of artificial intelligence in medicine. WIREs Data Mining and Knowledge Discovery, 9(4), e1312.
21. Goodfellow, I., Bengio, Y., & Courville, A. (2016). Deep learning. MIT Press.
22. Vaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017). Attention is all you need. Advances in Neural Information Processing Systems, 30, 5998–6008.
23. Zhang, Y., & Wallace, B. (2017). A sensitivity analysis of (and practitioners’ guide to) convolutional neural networks for sentence classification. Proceedings of the Eighth International Joint Conference on Natural Language Processing, 253–263.
24. Chen, T., & Guestrin, C. (2016). XGBoost: A scalable tree boosting system. Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, 785–794.
Downloads
Published
Issue
Section
License
Copyright (c) 2025 Journal of Advanced Artificial Intelligence Research
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.
