Prediction of Air Pollution with Machine Learning Algorithms

Year 2024, Volume: 19 Issue: 1, 1 - 12, 28.03.2024

Kahraman Oğuz , Muhammet Ali Pekin

https://doi.org/10.55525/tjst.1224661

Abstract

Air pollution has become an important problem due to its threats. Air pollutants are in complex interaction with atmosphere and environment. For this reason, it is important to study air pollution issues. In recent years, studies on prediction of air pollutants with machine learning methods have gained momentum. In this study, some air pollutants are predicted with various machine learning algorithms considering meteorological factors. In machine learning phase, a separate study is conducted with various machine learning algorithms (multilayer perceptron neural network, stochastic gradient descent, ridge regression, cross decomposition) considering temperature, relative humidity, wind, pressure and air pollutant measurements of previous hour. Consistencies of these algorithms in estimating pollutant concentrations are compared. Various statistical metrics are used to analyze the consistencies. As a result, the coefficient of determination of all algorithms are found above 0.67, considering the test section. It is found that the coefficient of determination of the multilayer perceptron neural network algorithm provides better results than other algorithms.

Keywords

Air pollution, machine learning, neural network, modeling, Çanakkale city

Hava Kirliliğinin Makine Öğrenme Algoritmaları ile Tahmin Edilmesi

Abstract

Hava kirliliği, canlı sağlığına yönelik tehditleri sebebiyle önemli bir problem haline gelmiştir. Hava kirliliği atmosfer ve çevre ile karmaşık ilişki içerisindedir. Bu nedenle hava kirliliği ile alakalı konuların çalışılması önemlidir. Son yıllarda hava kirleticilerinin makine öğrenmesi yöntemleriyle tahmin edilmesine yönelik çalışmalar hız kazanmıştır. Bu çalışmada, meteorolojik faktörler göz önüne alınarak çeşitli makine öğrenme algoritmaları ile bazı hava kirleticilerinin tahmini yapılmıştır. Makine öğrenmesi aşamasında, bir önceki saatin sıcaklık, bağıl nem, rüzgar, basınç ve hava kirletici ölçümleri dikkate alınarak çeşitli makine öğrenmesi algoritmaları (çok katmanlı algılayıcı sinir ağı, stokastik gradyan inişi, sırt regresyonu, çapraz ayrıştırma) ile ayrı ayrı çalışma yapılmıştır. Bu algoritmaların kirletici konsantrasyonlarını tahmin etmedeki tutarlılıkları karşılaştırılmıştır. Tutarlılıkları analiz etmek için çeşitli istatistiksel metrikler kullanılmıştır. Sonuç olarak, test bölümü dikkate alındığında tüm algoritmaların belirleme katsayısı 0.67'nin üzerinde bulunmuştur. Çok katmanlı algılayıcı sinir ağı algoritmasının belirleme katsayısının diğer algoritmalara göre daha iyi sonuçlar verdiği tespit edilmiştir.

Keywords

Hava kirliliği, makine öğrenmesi, sinir ağı, modelleme, Çanakkale ili

References

• Abed Al Ahad M, Sullivan F, Demšar U, Melhem M, Kulu H. The effect of air-pollution and weather exposure on mortality and hospital admission and implications for further research: A systematic scoping review. PloS one 2020; 15(10): e0241415.
• Hrdlickova Z, Michalek J, Kolar M, Vesely M. Identification of factors affecting air pollution by dust aerosol PM10 in Brno City, Czech Republic. Atmos Environ 2008; 42(37): 8661–8673.
• Ei-Sharkawy MF, Zaki GR. Effect of meteorological factors on the daily average levels of particulate matter in the Eastern Province of Saudi Arabia: a cross-sectional study. J Sci Technol 2015; 5(1): 18–29.
• Oğuz, K. Nevşehir İlinde Hava Kalitesinin ve Meteorolojik Faktörlerin Hava Kirliliği Üzerine Etkilerinin İncelenmesi. Doğal Afetler ve Çevre Dergisi 2020; 6(2): 391-404.
• Qin YG, Yi C, Dong GL, Min JZ. Investigating the influence of meteorological factors on particulate matters: A case study based on path analysis. Energy & Environment 2019; 31(3): 1-13.
• Panda N, Osthus D, Srinivasan G, O’Malley D, Chau V, Oyen D, Godinez H. Mesoscale informed parameter estimation through machine learning: A case-study in fracture modeling. Journal of Computational Physics 2020; 420: 1-15.
• Alpaydin E. Introduction to Machine Learning. The MIT Press: Cambridge, MA, USA, 2010.
• Gagliardi RV, Andenna C. A Machine Learning Approach to Investigate the Surface Ozone Behavior. Atmosphere 2020; 11(11): 1173.
• Kothandaraman D, Praveena N et al. Intelligent Forecasting of Air Quality and Pollution Prediction Using Machine Learning, Adsorption Science & Technology 2022; (5086622), 15.
• Kumar K, Pande BP. Air pollution prediction with machine learning: a case study of Indian cities. Int. J. Environ. Sci. Technol. 2022; 1-16.
• Ünaldı S, Yalçın N. Hava Kirliliğinin Makine Öğrenmesi Tabanlı Tahmini: Başakşehir Örneği. Mühendislik Bilimleri ve Araştırmaları Dergisi 2022; 4(1): 35-44.
• Oğuz K, Pekin MA. Makine Öğrenme Algoritmaları ile PM10 Konsantrasyon Tahmini. Journal of Advanced Research in Natural and Applied Sciences 2022; 8(2): 201-213.
• Bekkar A, Hssina B, Douzi S, Douzi K. Air-pollution prediction in smart city, deep learning approach. J Big Data 2021; 8(1): 161.
• Gültepe YA Comparative Assessment on Air Pollution Estimation by Machine Learning Algorithms. European Journal of Science and Technology 2019; (16): 8-15.
• Irmak ME, Aydilek İB. Using Ensemble Regression Algorithms for Improving the Prediction Success of Air Quality Index. Academic Platform Journal of Engineering and Science 2019; 7(3): 507-514.
• Dobrea M et al. Machine Learning algorithms for air pollutants forecasting. 2020 IEEE 26th International Symposium for Design and Technology in Electronic Packaging (SIITME) 2020; 109-113.
• Lee M, Lin L, Chen CY et al. Forecasting Air Quality in Taiwan by Using Machine Learning. Sci Rep 10, 2020; 4153.
• Doreswamy KSH, Yogesh KM, Gad I. Forecasting air pollution particulate matter (PM2.5) using machine learning regression models. Procedia Computer Science 2020; 171(2020): 2057-2066.
• Liang YC, Maimury Y, Chen AHL, Juarez JRC. Machine learning-based prediction of air quality. Appl. Sci. 10, 2020; 9151.
• Haykin S. Neural Networks: A Comprehensive Foundation. Prentice Hall: USA, 1999.
• Al-Saif AM, Abdel-Sattar M, Aboukarima AM, Eshra DH. Application of a multilayer perceptron artificial neural network for identification of peach cultivars based on physical characteristics. PeerJ 2021; 9(e11529).
• Gardner MW, Dorling SR, Artificial neural networks (the multilayer perceptron) —a review of applications in the atmospheric sciences. Atmospheric environment 1998; 32(14-15): 2627-2636.
• Latz J. Analysis of stochastic gradient descent in continuous time. Stat Comput. 2021; 31(39): 1-25.
• Orange Data Mining, 2022, Stochastic Gradient Descent, available at: https://orange3.readthedocs.io/projects/orange-visual-programming/en/latest/widgets/model/stochasticgradient.html (accessed: 1 June 2022).
• UFLDL Tutorial, 2022, Optimization: Stochastic Gradient Descent, available at: http://deeplearning.stanford.edu/tutorial/supervised/OptimizationStochasticGradientDescent/ (accessed: 15 June 2022).
• Ruder, 2016, An overview of gradient descent optimization algorithms, available at: https://ruder.io/optimizing-gradient-descent/#:~:text=Stochastic%20gradient%20descent,-Stochastic%20gradient%20descent&text=%CE%B8%3D%CE%B8%E2%88%92%CE%B7%E2%8B%85%E2%88%87%CE%B8J(%CE%B8%3B,%3B%20y%20(%20i%20)%20)%20 (accessed: 30 June 2022).
• Topal M, Eyduran E, Yağanoğlu AM, Sönmez A, Keskin S. Çoklu Doğrusal Bağlantı Durumunda Ridge ve Temel Bileşenler Regresyon Analiz Yöntemlerinin Kullanımı. Atatürk Üniversitesi Ziraat Fakültesi Dergisi 2013; 41(1): 53-57.
• Scikitlearn, 2022, Cross decomposition, available at: https://scikit-learn.org/stable/modules/cross_decomposition.html (accessed: 5 July 2022).
• ML-science, 2022, Cross decomposition, available at: https://www.ml-science.com/cross-decomposition (accessed: 5 July 2022).
• Singh, 2020, Understanding Data Preprocessing, available at: https://towardsdatascience.com/data-preprocessing-e2b0bed4c7fb (accessed: 28 July 2022).
• Baheti, 2022, A Simple Guide to Data Preprocessing in Machine Learning, available at: https://www.v7labs.com/blog/data-preprocessing-guide (accessed: 3 August 2022).
• DataTechNotes, 2019, Regression Model Accuracy (MAE, MSE, RMSE, R-squared) Check in R, available at: https://www.datatechnotes.com/2019/02/regression-model-accuracy-mae-mse-rmse.html (accessed: 10 August 2022).