Makine Öğrenimi Teknikleri ile Çevresel Adalet ve Hava Kalitesi Üzerine Bir Analiz

Öz

Çalışma, makine öğrenimi ve veri analizi yöntemlerini kullanarak hava kalitesini analiz etmeyi ve çevresel adalet ile hava kalitesine odaklanmayı amaçlamaktadır. Hızlı kentleşme, endüstriyel büyüme ve küresel çevresel zorluklarla birlikte hava kalitesi çalışmaları giderek daha fazla önem kazanmaktadır. Birleşmiş Milletler İklim Değişikliği Çerçeve Sözleşmesi, iklim değişikliği tanımını insan etkisini de kapsayacak şekilde genişletmiştir ve bu, yoğunlaşan iklim kriziyle yakından ilişkilidir. Bu çalışma, çeşitli ülkelerden elde edilen hava kalitesi verilerini analiz ederek hava kirliliği hakkında kapsamlı bir genel bakış sunmaktadır. Gelecekteki hava kalitesi eğilimlerini tahmin etmek ve sonuçlarını analiz ederek yorumlamak için Random Forest, Karar Ağaçları, XGBoost ve Adaboost gibi yöntemler kullanılarak makine öğrenimi modelleri geliştirilmiştir. Bu yöntemler arasında en yüksek tahmin performansını gösteren XGBoost modeli ile önümüzdeki 10 yıla ilişkin tahminler yapılmıştır. Bu tahminlere göre, Bhutan ve Kuzey Kore hava kalitesindeki en yüksek artışları gösterirken, Hindistan, Pakistan ve Nepal gibi ülkelerde belirgin düşüşler görülmüştür. Bu durum, hava kalitesindeki farklı eğilimleri yansıtmaktadır. Analiz, Laos, Endonezya ve Kuzey Kore'nin sırasıyla 0.183878, 0.116214 ve 0.114642 ile en kritik hava kalitesi değişimlerini yaşayacağını ortaya koymaktadır. Bu ülkeler, 2018 ile 2028 yılları arasında hava kalitesinde dikkate değer artışlar gösterecektir. Çalışma, çevresel adaletsizlik kavramını vurgulamakta ve karmaşık hava kalitesi verilerini anlaşılır bir şekilde sunmak için etkili veri görselleştirme tekniklerinden yararlanmaktadır.

Anahtar Kelimeler

• Hava Kalitesi Tahmini
• CO Seviyesi Tahmini
• Makine Öğrenimi

An Analysis on Environmental Justice and Air Quality Using Machine Learning Techniques

Öz

The study aims to analyze air quality using machine learning and data analysis methods, focusing on environmental justice and air quality. With rapid urbanization, industrial growth, and global environmental challenges, air quality studies are becoming increasingly important. The United Nations Framework Convention on Climate Change has broadened the definition of climate change to encompass human impact, and this closely links to the intensifying climate crisis. The study presents a comprehensive overview of air pollution by analyzing air quality data from various countries. We develop machine learning models using methodologies like Random Forest, Decision Tree, XGBoost, and Adaboost to predict future air quality trends and analyze and interpret their results. We made predictions for the next 10 years using the XGBoost model, which demonstrated the highest prediction performance among these methods. According to these predictions, Bhutan and North Korea have the highest increases, while countries such as India, Pakistan, and Nepal have noticeable decreases, reflecting diverse air quality trends. The analysis reveals that Laos, Indonesia, and North Korea will experience the most crucial changes in air quality, with changes of 0.183878, 0.116214, and 0.114642, respectively. These countries will have notable increases in their air quality from 2018 to 2028. The study emphasizes the concept of environmental injustice and uses effective data visualization techniques to visually present complex air quality data in an understandable manner.

Anahtar Kelimeler

• CO level prediction
• Air quality prediction
• machine learning

Kaynakça

1. [1] S. A. Aram, E. A. Nketiah, B. M. Saalidong, H. Wang, A. R. Afitiri, A. B. Akoto, and P. O. Lartey, “Machine learning-based prediction of air quality index and air quality grade: a comparative analysis,” International Journal of Environmental Science and Technology, vol. 21, no. 2, pp. 1345–1360, 2024.
2. [2] P. Bhalgat, S. Pitale, and S. Bhoite, “Air quality prediction using machine learning algorithms,” International Journal of Computer Applications Technology and Research, vol. 8, no. 9, pp. 367–370, 2019.
3. [3] Bhutan Electricity Authority, Bhutan energy statistics 2020. Thimphu, Bhutan: Bhutan Electricity Authority, 2020.
4. [4] L. Breiman, “Random forests,” Machine Learning, vol. 45, no. 1, pp. 5–32, 2001.
5. [5] M. Castelli, F. M. Clemente, A. Popovič, S. Silva, and L. Vanneschi, “A machine learning approach to predict air quality in California,” Complexity, 2020.
6. [6] T. Chen and C. Guestrin, “XGBoost: A scalable tree boosting system,” in Proc. 22nd ACM SIGKDD Int. Conf. on Knowledge Discovery and Data Mining, 2016, pp. 785–794.
7. [7] Y. Choe and H. Kim, “Emissions from small-scale industries in North Korea: A field study,” East Asian Industrial Review, vol. 3, no. 1, pp. 12–27, 2020.
8. [8] K. Dorji, P. Wangchuk, and T. Tshering, “Impact of hydropower construction on air quality in western Bhutan,” Journal of Himalayan Environmental Studies, vol. 5, no. 2, pp. 45–58, 2019.

9. [9] Y. Freund and R. E. Schapire, “A decision-theoretic generalization of on-line learning and an application to boosting,” Journal of Computer and System Sciences, vol. 55, no. 1, pp. 119–139, 1997.
10. [10] N. S. Gupta, Y. Mohta, K. Heda, R. Armaan, B. Valarmathi, and G. Arulkumaran, “Prediction of air quality index using machine learning techniques: a comparative analysis,” Journal of Environmental and Public Health, vol. 2023, no. 1, p. 4916267, 2023.
11. [11] M. Hardini, R. A. Sunarjo, M. Asfi, M. H. R. Chakim, and Y. P. A. Sanjaya, “Predicting air quality index using ensemble machine learning,” ADI Journal on Recent Innovation, vol. 5, no. 1Sp, pp. 78–86, 2023.
12. [12] T. Hastie, R. Tibshirani, and J. Friedman, The elements of statistical learning: Data mining, inference, and prediction, 2nd ed. Springer, 2009.
13. [13] International Energy Agency, World energy outlook 2021. Paris, France: IEA, 2021.
14. [14] D. S. Kim and J. Y. Park, “Household energy use and indoor air quality in rural DPRK,” Journal of Asian Public Health, vol. 10, no. 4, pp. 210–222, 2019.
15. [15] S. H. Lee, M. J. Kang, and H. J. Park, “Satellite-based assessment of coal consumption in North Korea,” International Journal of Remote Sensing, vol. 39, no. 14, pp. 4768–4782, 2018.
16. [16] T. M. Lei, S. W. Siu, J. Monjardino, L. Mendes, and F. Ferreira, “Using machine learning methods to forecast air quality: A case study in Macao,” Atmosphere, vol. 13, no. 9, p. 1412, 2022.
17. [17] Q. Liu, B. Cui, and Z. Liu, “Air quality class prediction using machine learning methods based on monitoring data and secondary modeling,” Atmosphere, vol. 15, no. 5, p. 553, 2024.
18. [18] Ministry of Energy and Mineral Resources, Annual energy report 2020: Fuel consumption and emissions. Jakarta, Indonesia, 2020.
19. [19] Ministry of Environment, Bhutan, Policy on single-use plastics and electric vehicle promotion. Thimphu, Bhutan, 2019.
20. [20] J. H. Park, M. K. Lee, and S. Y. Choi, “Evaluating public transportation emissions in Pyongyang,” Journal of Urban Environmental Studies, vol. 7, no. 2, pp. 101–113, 2020.
21. [21] T. Phuntsho and Y. Tashi, “Transboundary air pollution from India to Bhutan: An analysis,” Environmental Science and Policy Journal, vol. 12, no. 1, pp. 23–34, 2020.
22. [22] J. R. Quinlan, “Induction of decision trees,” Machine Learning, vol. 1, no. 1, pp. 81–106, 1986.
23. [23] G. Ravindiran, G. Hayder, K. Kanagarathinam, A. Alagumalai, and C. Sonne, “Air quality prediction by machine learning models: A predictive study on the Indian coastal city of Visakhapatnam,” Chemosphere, vol. 338, p. 139518, 2023.
24. [24] United Nations Environment Programme, Air pollution in South Asia: Transboundary haze report. Nairobi, Kenya, 2020.
25. [25] S. Wang, J. McGibbon, and Y. Zhang, “Predicting high-resolution air quality using machine learning: Integration of large eddy simulation and urban morphology data,” Environmental Pollution, vol. 344, p. 123371, 2024.
26. [26] World Bank, World Development Indicators: Transport statistics. Washington, DC: World Bank, 2021.
27. [27] Y. Zhang, Y. Wang, M. Gao, Q. Ma, J. Zhao, R. Zhang, and L. Huang, “A predictive data features exploration-based air quality prediction approach,” IEEE Access, vol. 7, pp. 30732–30743, 2019.