IJEP 45(6): 487-499 : Vol. 45 Issue. 6 (June 2025)
1. Chandigarh College of Engineering and Technology, Computer Science and Engineering, Chandigarh – 160 019, Haryana, India
2. Chandigarh College of Engineering and Technology, Electronics and Communication Engineering, Chandigarh – 160 019, Haryana, India
Abstract
The quality of air, an essential natural resource, has been compromised due to economic activities. A major issue in this modern world is air pollution, which is caused by pollutant gases from various vehicles, factories and coal burning. Much research has been done on live and accurate PM2.5 pollutant prediction by scientists and researchers. In this paper, data is obtained from the Central Pollution Control Board (CPCB) and a prediction model is developed for air pollutant prediction. This paper uses a transit search optimization tunned DnCNN algorithm with machine learning methods to predict air pollution deposition on tree leaves in high-medium-low traffic areas. The accuracy is 85% for linear regression, 89% for k-nearest neighbours (KNN), 87% for polynomial regression and 91% for support vector machine (SVM).
Keywords
Air pollutant, Transit search optimization, Machine learning, PM2.5, Prediction, Hyperparameter tuning
References
- Chen, L.J., et al. 2017. An open framework for participatory PM2.5monitoring in smart cities. IEEE Access. 5: 14441–14454. DOI: 10.1109/ACCESS. 2017.2723919.
- WHO. 2021. Health impacts. World Health Organization, Geneva. Available at: https://www. who.int/teams/environment-climate-change-and-health/air-quality-and-health/health-impacts. Accessed on 28 May 2022.
- Hooyberghs, J., et al. 2005. A neural network forecast for daily average PM10concentrations in Belgium. Atmos. Env., 39: 3279–3289. DOI: 10. 1016/j.atmosenv.2005.01.050.
- Bergstra, J., et al. 2011. Algorithms for hyperpara-meter optimization. In Advances in neural information processing systems. Ed K.Q. Weinberger, et al. Curran Associates, Inc.
- Gawronski, S.W., et al. 2016. Plants in air phyto-remediation (chapter 8). In Advances in botanical research (vol 3). Ed A. Cuypers and J. Vangrons-veld. Academic Press, Cambridge, MA, USA. 83: 319–346. DOI: 10.1016/bs.abr. 2016.12.008.
- Klingberg, J., et al. 2017. Influence of urban vegetation on air pollution and noise exposure- A case study in Gothenburg, Sweden. Sci. Total Env., 599–600: 1728–1739. DOI: 10.1016/j.scitotenv.2017. 05.051.
- Jeanjean, A.P.R., et al. 2017. Air quality affected by trees in real street canyons: The case of Maryle-bone neighbourhood in central London. Urban Urban Green. 22: 41–53. DOI: 10.1016/j.ufug.2017. 01.009.
- Nowak, D.J., et al. 2022. Tree and forest effects on air quality and human health in the United States. Env. Poll., 193: 119–129. DOI: 10.1016/j.envpol. 2014.05.028.
- Mirsanjari, M.M., et al. 2020. Investigation of the impacts of urban vegetation loss on the ecosystem service of air pollution mitigation in Karaj metropolis, Iran. Env. Monit. Assess., 192(8): 501. DOI: 10.1007/s10661-020-08399-8.
- Xing, Y. and P. Brimblecombe. 2019. Role of vegetation in deposition and dispersion of air pollution in urban parks. Atmos. Env., 201: 73–83. DOI: 10.1016/j.atmosenv.2018. 12.027.
- Nemitz, E., et al. 2020. Potential and limitation of air pollution mitigation by vegetation and uncertainties of deposition-based evaluations. Philos. Trans. Math. Phys. Eng. Sci., 378: 20190320. DOI: 10.1 098/rsta.2019.0320.
- Abreu, S. 2019. Automated architecture design for deep neural networks. Bachelor Thesis. Jacobs University, Bremen, Germany.
- Lerman, P.M. 1980. Fitting segmented regression models by grid search. J. Royal Statistical Society. Series C : Appl. Statistics. 29(1): 77-84.
- Bergstra, J. and Y. Bengio. 2012. Random search for hyper-parameter optimization. J. Machine Learning Res., 13: 281-305.
- Strobl, C., et al. 2008. Conditional variable importance for random forests. BMC Bioinfor., 9: 1–11.
- Liaw, R., et al. 2018. Tune: a research platform for distributed model selection and training. Arxiv. DOI: 10.48550/arXiv.1807.05118.
- Boussaïd, I., J. Lepagnot and P. Siarry. 2013. A survey on optimization meta-heuristics. Inf. Sci., 237: 82–117.
- Brewer, T.L. 2019. Black carbon emissions and regulatory policies in transportation. Energy Policy. 129 (March): 1047–1055. DOI: 10.1016/j. enpol. 2019.02.073.
- WHO. 2013. Review of evidence on health aspects of air pollution: REVIHAAP project: Technical report. World Health Organization, Geveva.
- De Miranda, R.M., et al. 2019. Relationship between black carbon (BC) and heavy traffic in Sao Paulo, Brazil. Transportation Res. Part D: Transport Env., 68(September): 84–98. DOI: 10.1016/j.trd.2017.09.002.
- Hoek, G., et al. 2008. A review of landuse regression models to assess spatial variation of outdoor air pollution. Atmos. Env., 42(33): 7561-7578. DOI: 10.1016/j. atmosenv.2008.05.057.
- Park, S., et al. 2018. Predicting PM10concentration in Seoul metropolitan subway stations using artificial neural network (ANN). J. Hazard. Mater., 341: 75–82.
- Rybarczyk, Y. and R. Zalakeviciute. 2018. Machine learning approaches for outdoor air quality modelling: A systematic review. Appl. Sci., 8: 2570.
- Barida, A.J. and H. Hadiyantob. 2024. Hyper para-meter optimization for hourly PM2.5pollutant prediction. J. Emerg. Sci. Eng., 2(1): e15. DOI: 10.614 35/jese.2024.e15.
- Yang, F. and J. Xiaoming. 2023. Air quality index prediction model based on multiple attention mechanisms and hyper parameter optimization. 3rd International Conference on Electronic information engineering and computer science (EIECS). Institute Electrical Electronics Engineers (IEEE). DOI: 10.110 9/EIECS59936.2023.10435483.
- Das, S. and A. Alzimami. 2023. Bayesian optimization based hyperparameter tuning of ensemble regression models in smart city air quality monitoring data analytics. Institute Electrical Electronics Engineers (IEEE). DOI: 10.1109/icaisc56366.2023.10 085504.
- Kusmahendra, A., et al. 2022. System design for air quality monitoring and classification using artificial neural network with hyperparameters optimization. international electronics symposium (IES).
- Ahmad, T., P. Purwono and G. Windu. 2022. Air quality prediction model with support vector machines with grid search hyperparameter optimization CV. BuletinIlmiah SarjanaTeknik Elektro. 4(1): 12-21. DOI: 10.12928/biste.v4i1.
- Yin, P., et al. 2019. Ensemble learning of model hyper parameters and spatio-temporal data for calibration of low-cost PM2.5sensors. Mathematical Biosci. Eng., 16(6): 6858-6873. DOI: 10.3934/mbe. 2019343.
- Songa, Y., C. Zhanga and X. Jin. 2023. Spatial prediction of PM2.5concentration using hyperparameter optimization XGBoost model in China. Env. Tech. Innov., 32: 103272. DOI: 10.1016/j.eti.2023.103 272.
- Percival, J., et al. 2023. Indoor pollutant classification modelling using relevant sensors under thermodynamic conditions with multi-layer perceptron hyperparameter tuning. Int. J. Adv. Computer Sci. Applications. 14: 2.
- Hamida, S., et al. 2020. Optimization of machine learning algorithms hyperparameters for improving the prediction of patients infected with Covid-19. 2nd International Conference on Electronics, control, optimization and computer science (ICECOCS). IEEE. DOI: 10.1109/ICECOCS50124. 2020.931 4373.