Skip to main content
SpringerLink
Log in

Probabilistic air quality forecasting using deep learning spatial–temporal neural network

  • Published:
GeoInformatica Aims and scope Submit manuscript

Abstract

Regional air quality monitoring, a critical component of sustainable development is realized through various air quality observation stations established across a region. Accurate forecasting of air quality data collected from these observation stations requires the modelling of spatial–temporal patterns in the data. Deep learning algorithms, known for their ability to capture layers of abstraction, can proficiently achieve spatial–temporal modeling. However, deterministic models that produces point forecast does not consider the underlying model uncertainty during prediction and are therefore less reliable for real-time applications. Probabilistic forecasting models that forecast prediction intervals rather than point estimates can overcome this through uncertainty quantification. The objective of the proposed study is three-fold: i) develop an efficient deterministic deep learning spatial–temporal neural network named DL-STNN for spatial–temporal air quality forecasting; ii) investigate different approaches to uncertainty quantification in deep learning models and integrate some of them, such as Monte-Carlo Dropout, Ensemble Averaging, Gaussian Process Regression, Quantile Regression, and Bayesian Inference, in tandem with DL-STNN to facilitate probabilistic forecasting; and iii) evaluate the developed deterministic and probabilistic models, using a real-world Delhi air quality dataset. The evaluation results show that, among the deterministic models, DL-STNN outperforms the baselines with 39.8% more accurate predictions and performs consistently across all seasons in Delhi. Furthermore, among the DL-STNN-based tandem models that performed probabilistic forecasting, Bayesian DL-STNN proved efficient. It does 13% more accurate point forecasting and has 20% higher suitability score than the other tandem models, indicating that Bayesian inference adapts DL-STNN more reliable for real-time applications.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Algorithm 1:
Algorithm 2:
Algorithm 3:
Algorithm 4:
Algorithm 5:
Algorithm 6:
Fig. 7
Fig. 8
Fig. 9
Fig. 10
Fig. 11

Similar content being viewed by others

Data availability

The datasets generated during and/or analysed during the current study are available from the corresponding author on reasonable request.

Reference s

  1. Abirami S, Chitra P (2019) Real Time Twitter Based Disaster Response System for Indian Scenarios. In: 2019 26th International Conference on High Performance Computing, Data and Analytics Workshop (HiPCW). IEEE, pp 82–86

  2. Abirami S, Chitra P (2022) Regional spatio-temporal forecasting of particulate matter using autoencoder based generative adversarial network. Stoch Environ Res Risk Assess 1–22

  3. Abirami S, Chitra P (2021) Regional air quality forecasting using spatiotemporal deep learning. J Clean Prod 283:125341

    Article  Google Scholar 

  4. Abirami S, Chitra P, Madhumitha R, Kesavan SR (2020) Hybrid Spatio-temporal Deep Learning Framework for Particulate Matter(PM2.5) Concentration Forecasting. In: 2020 International Conference on Innovative Trends in Information Technology (ICITIIT). pp 1–6

  5. Akbal Y, Ünlü KD (2021) A deep learning approach to model daily particular matter of Ankara: key features and forecasting. Int J Environ Sci Technol 1–17

  6. Al-Shedivat M, Wilson AG, Saatchi Y et al (2017) Learning scalable deep kernels with recurrent structure. J Mach Learn Res 18:2850–2886

    MathSciNet  MATH  Google Scholar 

  7. Altikat S (2021) Prediction of CO 2 emission from greenhouse to atmosphere with artificial neural networks and deep learning neural networks. Int J Environ Sci Technol 1–10

  8. Arora N (2021) New Delhi is world’s most polluted capital for third straight year - IQAir study | Reuters. https://www.reuters.com/article/us-india-pollution-idUSKBN2B817F. Accessed 29 Nov 2021

  9. Atencia M, Stoean R, Joya G (2020) Uncertainty quantification through dropout in time series prediction by echo state networks. Mathematics 8:1374

    Article  Google Scholar 

  10. Atluri G, Karpatne A, Kumar V (2018) Spatio-temporal data mining: A survey of problems and methods. ACM Comput Surv 51:1–41

    Article  Google Scholar 

  11. Benhaddi M, Ouarzazi J (2021) Multivariate time series forecasting with dilated residual convolutional neural networks for urban air quality prediction. Arab J Sci Eng 46:3423–3442

    Article  Google Scholar 

  12. Blundell C, Cornebise J, Kavukcuoglu K, Wierstra D (2015) Weight uncertainty in neural network. In: International Conference on Machine Learning. PMLR, pp 1613–1622

  13. Chandra Mouli V, Chitra P, Harihara Subramanian M, Abirami S (2022) A deep learning ensemble model for short-term rainfall prediction. In: 2022 International Conference on Wireless Communications Signal Processing and Networking (WiSPNET), pp 135–138. https://doi.org/10.1109/WiSPNET54241.2022.9767163

  14. Chaudhary V, Deshbhratar A, Kumar V, Paul D (2018) Time series based LSTM model to predict air pollutant’s concentration for prominent cities in India. UDM, Aug

  15. Chen T, Qian Z, Jing B, et al (2020a) Probabilistic Wind Speed Forecasting based on Minimal Gated Unit and Quantile Regression. In: Journal of Physics: Conference Series. IOP Publishing, p 12039

  16. Chen Y, Kang Y, Chen Y, Wang Z (2020) Probabilistic forecasting with temporal convolutional neural network. Neurocomputing 399:491–501

    Article  Google Scholar 

  17. Chitra P, Abirami S (2019) Smart pollution alert system using machine learning. In: Integrating the internet of things into software engineering practices. IGI Global, pp 219–235

  18. Chitra P, Abirami S (2020) Leveraging fog computing and deep learning for building a secure individual health-based decision support system to evade air pollution. In: Security, privacy, and forensics issues in big data. IGI Global, pp 380–406

  19. Dadhich AP, Goyal R, Dadhich PN (2018) Assessment of spatio-temporal variations in air quality of Jaipur city, Rajasthan, India. Egypt J Remote Sens Sp Sci 21:173–181

    Google Scholar 

  20. Du S, Li T, Yang Y, Horng S-J (2019) Deep Air Quality Forecasting Using Hybrid Deep Learning Framework. IEEE Trans Knowl Data Eng 1–1. https://doi.org/10.1109/tkde.2019.2954510

  21. Express Web Desk (2016) Diwali effect: Pollution worsens, particulate matter soars in Delhi. In: The Indian Express. http://indianexpress.com/article/india/india-news-india/post-diwali-pm-shoots-up-10-times-more-than-the-safe-limit-3730200/

  22. Feng R, Zheng H jun, Gao H et al (2019) Recurrent Neural Network and random forest for analysis and accurate forecast of atmospheric pollutants: A case study in Hangzhou, China. J Clean Prod 231:1005–1015. https://doi.org/10.1016/j.jclepro.2019.05.319

    Article  Google Scholar 

  23. Fortunato M, Blundell C, Vinyals O (2017) Bayesian recurrent neural networks. arXiv Prepr arXiv170402798

  24. Freeman BS, Taylor G, Gharabaghi B, Thé J (2018) Forecasting air quality time series using deep learning. J Air Waste Manage Assoc 68:866–886

    Article  Google Scholar 

  25. Gal Y, Ghahramani Z (2016) Dropout as a bayesian approximation: Representing model uncertainty in deep learning. In: international conference on machine learning. PMLR, pp 1050–1059

  26. Gu K, Qiao J, Lin W (2018) Recurrent air quality predictor based on meteorology-and pollution-related factors. IEEE Trans Ind Informatics 14:3946–3955

    Article  Google Scholar 

  27. Hernández-Lobato JM, Adams R (2015) Probabilistic backpropagation for scalable learning of bayesian neural networks. In: International conference on machine learning. PMLR, pp 1861–1869

  28. Hochreiter S, Schmidhuber J (1997) Long short-term memory. Neural Comput 9:1735–1780

    Article  Google Scholar 

  29. Huang CJ, Kuo PH (2018) A deep cnn-lstm model for particulate matter (Pm2.5) forecasting in smart cities. Sensors (Switzerland) 18. https://doi.org/10.3390/s18072220

  30. Kambezidis HD, Tulleken R, Amanatidis GT et al (1995) Statistical evaluation of selected air pollutants in Athens, Greece. Environmetrics 6:349–361

    Article  Google Scholar 

  31. Kang GK, Gao JZ, Chiao S et al (2018) Air quality prediction: Big data and machine learning approaches. Int J Environ Sci Dev 9:8–16

    Article  Google Scholar 

  32. Kersting K, Plagemann C, Pfaff P, Burgard W (2007) Most likely heteroscedastic Gaussian process regression. In: Proceedings of the 24th international conference on Machine learning. pp 393–400

  33. Kingma DP, Salimans T, Welling M (2015) Variational dropout and the local reparameterization trick. Adv Neural Inf Process Syst 28:2575–2583

    Google Scholar 

  34. Kingma DP, Welling M (2019) An introduction to variational autoencoders. arXiv Prepr arXiv190602691

  35. Kirch W (ed) (2008) Pearson’s Correlation Coefficient. In: Encyclopedia of Public Health. Springer Netherlands, Dordrecht, pp 1090–1091

  36. Koenker R, Hallock KF (2001) Quantile regression. J Econ Perspect 15:143–156

    Article  Google Scholar 

  37. Koh PW, Liang P (2017) Understanding black-box predictions via influence functions. In: International Conference on Machine Learning. PMLR, pp 1885–1894

  38. Korattikara A, Rathod V, Murphy K, Welling M (2015) Bayesian dark knowledge. arXiv Prepr arXiv150604416

  39. Krishnappa L, Devatha CP (2019) Machine Learning Approaches for the Estimation of Particulate Matter (PM2.5) Concentration Levels: A Case Study in the Hyderabad City, India. pp 765–774

  40. Lee S, Shin J (2019) Hybrid model of convolutional LSTM and CNN to predict particulate matter. Int J Inf Electron Eng 9:34–38. https://doi.org/10.18178/ijiee.2019.9.1.701

    Article  Google Scholar 

  41. Lu X, Cui X (2020) A spatiotemporal neural network modeling method for nonlinear distributed parameter systems. IEEE Trans Ind Informatics 17:1916–1926

    Article  Google Scholar 

  42. Lu X, Zou W, Huang M (2016) A novel spatiotemporal LS-SVM method for complex distributed parameter systems with applications to curing thermal process. IEEE Trans Ind Informatics 12:1156–1165

    Article  Google Scholar 

  43. Ma J, Ding Y, Cheng JCP et al (2019) A temporal-spatial interpolation and extrapolation method based on geographic Long Short-Term Memory neural network for PM2. 5. J Clean Prod 237:117729

    Article  Google Scholar 

  44. Ma X, Dai Z, He Z et al (2017) Learning traffic as images: a deep convolutional neural network for large-scale transportation network speed prediction. Sensors 17:818

    Article  Google Scholar 

  45. Mahalingam U, Elangovan K, Dobhal H, et al (2019) A machine learning model for air quality prediction for smart cities. In: 2019 International Conference on Wireless Communications Signal Processing and Networking (WiSPNET). pp 452–457

  46. Mahmoud A, Zrigui M (2021) BLSTM-API: Bi-LSTM recurrent neural network-based approach for Arabic paraphrase identification. Arab J Sci Eng 46:4163–4174

    Article  Google Scholar 

  47. McDermott PL, Wikle CK (2017) An ensemble quadratic echo state network for non-linear spatio-temporal forecasting. Stat 6:315–330

    Article  MathSciNet  Google Scholar 

  48. Mokhtari I, Bechkit W, Rivano H, Yaici MR (2021) Uncertainty-aware deep learning architectures for highly dynamic air quality prediction. IEEE Access 9:14765–14778. https://doi.org/10.1109/ACCESS.2021.3052429

    Article  Google Scholar 

  49. Mwangi B, Tian TS, Soares JC (2014) A review of feature reduction techniques in neuroimaging. Neuroinformatics 12:229–244

    Article  Google Scholar 

  50. Niska H, Hiltunen T, Karppinen A et al (2004) Evolving the neural network model for forecasting air pollution time series. Eng Appl Artif Intell 17:159–167

    Article  Google Scholar 

  51. Parveen N, Siddiqui L, Sarif MDN et al (2021) Industries in Delhi: Air Pollution versus Respiratory Morbidities. Process Saf Environ Prot

  52. Qi Y, Li Q, Karimian H, Liu D (2019) A hybrid model for spatiotemporal forecasting of PM2. 5 based on graph convolutional neural network and long short-term memory. Sci Total Environ 664:1–10

    Article  Google Scholar 

  53. Sakarkar G, Pillai S, Rao CV et al (2020) Comparative Study of Ambient Air Quality Prediction System Using Machine Learning to Predict Air Quality in Smart City. pp 175–182

  54. Seedat N, Kanan C (2019) Towards calibrated and scalable uncertainty representations for neural networks. arXiv Prepr arXiv191100104

  55. Seinfeld JH (1975) Air pollution: physical and chemical fundamentals. McGraw-Hill Companies

  56. Shanthini KM, Chitra P, Abirami S et al (2021) Recommendation of product value by extracting expiry date using deep neural network. In: 2021 12th International Conference on Computing Communication and Networking Technologies (ICCCNT). IEEE, pp 1–7

  57. Shepard D (1968) Proceedings of the 1968 23rd ACM National conference

  58. Shridhar K, Laumann F, Liwicki M (2018) Uncertainty estimations by softplus normalization in bayesian convolutional neural networks with variational inference. arXiv Prepr arXiv180605978

  59. Soh PW, Chang JW, Huang JW (2018) Adaptive deep learning-based air quality prediction model using the most relevant spatial-temporal relations. IEEE Access 6:38186–38199. https://doi.org/10.1109/ACCESS.2018.2849820

    Article  Google Scholar 

  60. Sriram S, Dwivedi AK, Chitra P, et al (2022) DeepComp: A Hybrid Framework for Data Compression Using Attention Coupled Autoencoder. Arab J Sci Eng 1–16

  61. Wang HW, Li XB, Wang D et al (2020) Regional prediction of ground-level ozone using a hybrid sequence-to-sequence deep learning approach. J Clean Prod 253:1–12. https://doi.org/10.1016/j.jclepro.2019.119841

    Article  Google Scholar 

  62. Wu W, Chen K, Qiao Y, Lu Z (2016) Probabilistic short-term wind power forecasting based on deep neural networks. In: 2016 International Conference on Probabilistic Methods Applied to Power Systems (PMAPS). IEEE, pp 1–8

  63. Zhan X, Qin H, Liu Y et al (2020) Variational Bayesian neural network for ensemble flood forecasting. Water 12:2740

    Article  Google Scholar 

  64. Zhang J, Yan J, Infield D et al (2019) Short-term forecasting and uncertainty analysis of wind turbine power based on long short-term memory network and Gaussian mixture model. Appl Energy 241:229–244

    Article  Google Scholar 

  65. Zhang L, Na J, Zhu J et al (2021) Spatiotemporal causal convolutional network for forecasting hourly PM2. 5 concentrations in Beijing. China. Comput Geosci 155:104869

    Article  Google Scholar 

  66. 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. https://doi.org/10.1016/j.chemosphere.2018.12.128

    Article  Google Scholar 

  67. Zhou Y, Chang F-J, Chang L-C et al (2019) Explore a deep learning multi-output neural network for regional multi-step-ahead air quality forecasts. J Clean Prod 209:134–145

    Article  Google Scholar 

  68. Zhu A, Wu Q, Cui R et al (2020) Exploring a rich spatial–temporal dependent relational model for skeleton-based action recognition by bidirectional LSTM-CNN. Neurocomputing 414:90–100

    Article  Google Scholar 

  69. Zhu D, Cai C, Yang T, Zhou X (2018) A machine learning approach for air quality prediction: model regularization and optimization. Big Data Cogn Comput 2:5. https://doi.org/10.3390/bdcc2010005

    Article  Google Scholar 

  70. Zhu Q, Chen J, Shi D et al (2020) Learning temporal and spatial correlations jointly: a unified framework for wind speed prediction. IEEE Trans Sustain Energy 11:509–523. https://doi.org/10.1109/TSTE.2019.2897136

    Article  Google Scholar 

  71. Zhu S, Yuan X, Xu Z et al (2019) Gaussian mixture model coupled recurrent neural networks for wind speed interval forecast. Energy Convers Manag 198:111772

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

  1. School of Computer Science and Engineering, Vellore Institute of Technology, Chennai, 600127, India

    S. Abirami

  2. Department of Computer Science Engineering, Thiagarajar College of Engineering, Madurai, 625015, India

    P. Chitra

Authors
  1. S. Abirami
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. P. Chitra
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to S. Abirami.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Additional information

Publisher's note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Appendix

Appendix

Figure 12

Figure 13

Fig. 12
figure 12

Model Architecture of DL-STNN-GPR

Full size image
Fig. 13
figure 13

Model Architecture of DL-STNN-QR

Full size image

Rights and permissions

Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Abirami, S., Chitra, P. Probabilistic air quality forecasting using deep learning spatial–temporal neural network. Geoinformatica 27, 199–235 (2023). https://doi.org/10.1007/s10707-022-00479-w

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10707-022-00479-w

Keywords

Search

Navigation