Air Pollution Modelling with Deep Learning: A Review

Abstract

Air pollution is one of the fundamental environmental problems of the industrialized world due to its adverse effects on all organisms. Several institutions warn that there exist serious air pollution in many regions of the world. When all devastating effects of air pollutants considered, it is crucial to create valid models to predict air pollution levels in order to determine future concentrations or to locate pollutant sources. These models may provide policy implications for governments and central authorities in order to prevent the excessive pollution levels. Though there are a number of attempts to model pollution levels in the literature, recent advances in deep learning techniques are promising more accurate prediction results along with integration of more data.  In this study, a detailed research about modelling with deep learning architectures on real air pollution data is given. With the help of this research we attempt to develop air pollution architectures with deep learning in future and enhance the results further with insights from recent advances of deep learning research such as Generative Adversarial Networks (GANs), where two competing networks are working against each other, one for creating a more realistic data and the other one to predict the state.   

Keywords

• Deep learning,air pollution estimation,artificial neural networks,generative adversarial model

References

1. [1] Copeland, M., 2016, What’s the Difference Between Artificial Intelligence, Machine Learning, and Deep Learning?, https://blogs.nvidia.com/blog/2016/07/29/whats-difference-artificial-intelligence-machine-learning-deep-learning-ai/, retrieval date: 24.04.2018.
2. [2] Şeker, A., Diri, B., Balık, H.H., 2017, Derin Öğrenme Yöntemleri ve Uygulamaları Hakkında Bir İnceleme, Gazi Mühendislik Bilimleri Dergisi 2017, 3(3): 47-64.
3. [3] Wang, S.C., 2003, Artificial Neural Network, The Springer International Series in Engineering and Computer Science, Volume 743, 81-100.
4. [4] Alsugair, A. M., Al-Qudrah, A. A. 1998, Artificial neural network approach for pavement maintenance, J. Comput. Civil Eng. ASCE, 2 (4), 249–255.
5. [5] Sarle, W. 1997, Neural network frequently asked questions, ftp://ftp.sas.com/pub/neural/FAQ.html, retrieval date: 07.03.2018.
6. [6] Kök, İ., Şimşek, M.U., Özdemir, S., 2017, A deep learning model for air quality prediction in smart cities, 2017 IEEE International Conference on Big Data (BIGDATA), 1973-1980.
7. [7] Reddy, V., Yedavalli, P., Mohanty, S., Nakhat, U., 2017, Deep Air: Forecasting Air Pollution in Beijing, China, https://www.ischool.berkeley.edu/sites/default/files/sproject_attachments/deep-air-forecasting_final.pdf, retrieval date: 25.04.2018.
8. [8] Li, X., Peng, L., Hu, Y., Shao, J., Chi, T., 2016, Deep learning architecture for air qualitypredictions, Environmental Science and Pollution Research.

9. [9] Qi, Z., Wang, T., Song, G., Hu, W., Li, X., Zhang, Z.M., 2018, Deep Air Learning: Interpolation, Prediction, and Feature Analysis of Fine-grained Air Quality, IEEETransactions on Knowledge and Data Engineering, https://ieeexplore.ieee.org/stamp/stamp.jsp?tp=&arnumber=8333777, retrieval date: 20.04.2018.
10. [10] Zhang, C., Yan, Z., Li, C., Rui, X., Liu, L., Bie, R., 2016, On Estimating Air Pollution from Photos Using Convolutional Neural Network, Proceedings of the 2016 ACM on Multimedia Conference, 297-301.
11. [11] Bui, T.C., Le, V.D., Cha, S.K. 2018. A Deep Learning Approach for Forecasting Air Pollution in South Korea Using LSTM, https://arxiv.org/abs/1804.07891, retrieval date: 22.04.2018.