Home / Current Issue / Paper 1717878
An Intelligent Air Quality Forecasting System Using Historical Data and Machine Learning
Subject area: Physical Sciences and Environment · Area of research: Environmental Science and Machine Learning
DOI: https://doi.org/10.64388/IREV9I11-1717878
Abstract
Air pollution is a significant environmental problem in urban areas, impacting millions of people's lives and leading to fatal and serious health effects. The World Health Organization (WHO) reports that almost 7 million people die from air pollution every year, and 90% of the world's population is breathing unhealthy air. While the Air Quality Index (AQI) helps represent air pollution levels in a standardised manner, most monitoring systems do not have any predictive ability for taking proactive decisions. This research introduces an intelligent air quality forecasting system with historical data and machine learning techniques to forecast the future air quality level. The system takes data from the major Indian cities from a multi-year period and incorporates certain pollutants (PM2.5, PM10, NO₂, SO₂, CO, O₃) and meteorological parameters. The data is preprocessed to deal with missing and inconsistent data. Various models such as regression models, ensemble models and deep learning models are tested. The results indicate that the advanced models are more accurate than the traditional ones, with XGBoost reaching an accuracy of more than 95% and the Bi-LSTM being able to capture temporal patterns well. The system provides 24-, 48-and 72-hour forecasts, which can be used for short term or medium-term planning. Particulate matter and meteorological factors are identified as prominent features in feature analysis. In general, it is scalable, and can help in making informed decisions for better air quality management and public health.
Keywords
Air Quality Index (AQI), Machine Learning, Time Series Forecasting, XGBoost, LSTM, Particulate Matter (PM2.5/PM10), Air Pollution, Meteorological Parameters, Public Health, Ensemble Learning, India
References
[1] M. S. Maheswari, D. Roshni, R. N. S, and S. S, “Enhancing Air Quality Forecasting Using Bi-LSTM: An AI-Driven Approach for Particulate Matter Prediction,” in Proc. 2nd Int. Conf. Computing and Data Science (ICCDS), 2025.
[2] Y. Liu, W. Cao, Y. Liu, D. Li, and Q. Wang, “Ensemble Online Sequential Extreme Learning Machine for Air Quality Prediction,” in Proc. IEEE Int. Conf. Control Science and Systems Engineering (ICCSSE), 2021, pp. 233–237.
[3] L. Zhang, W. Cai, K. Xing, H. Kambara, and W. Cai, “Monitoring and Evaluation of Air Quality in Jinan Based on Machine Learning Random Forest Model,” in Proc. Int. Symp. Computer Applications and Information Technology (ISCAIT), 2025.
[4] P. M. Papitha, J. J. B. Jayachandran, and B. S, “Predictive Modeling for Air Quality: A Machine Learning System,” in Proc. Int. Conf. Data Science, Agents and Artificial Intelligence (ICDSAAI), 2023.
[5] K. M. O. V. K. Kekulanadara, B. T. G. S. Kumara, and B. Kuhaneswaran, “Machine Learning Approach for Predicting Air Quality Index,” in Proc. Int. Conf. Decision Aid Sciences and Applications (DASA), 2021, pp. 622–626.
[6] I. W. A. Suranata, S. Basuki, K. A. A. Aryanto, P. A. W. Santiary, I. K. Swardika, and I. N. K. Wardana, “Federated Learning Approach for Air Quality Classification in Indonesia,” in Proc. Int. Conf. Smart Computing and Communication (ICSCC), 2024.
[7] M. Herath, H. Dutta, R. Minerva, N. Crespi, M. Alvi, and S. M. Raza, “An Integrated Digital Twin Architecture for Real-Time Urban Air Quality Management,” in Proc. IEEE/IFIP Int. Conf. Dependable Systems and Networks Workshops (DSN-W), 2025.
[8] V. R. Pasupuleti, Uhasri, P. Kalyan, S. Srikanth, and H. K. Reddy, “Air Quality Prediction of Data Log by Machine Learning,” in Proc. Int. Conf. Advanced Computing and Communication Systems (ICACCS), 2020, pp. 1395–1399.
[9] T. M. Amado and J. C. Dela Cruz, “Development of Machine Learning-Based Predictive Models for Air Quality Monitoring and Characterization,” in Proc. IEEE Region 10 Conf. (TENCON), 2018, pp. 668–672.
[10] A. Y. Prinanto, “Nova PM Sensor SDS011 for Alternative Air Quality Monitoring Based on the Internet of Things,” in Proc. Int. Conf. Adisutjipto on Aerospace Electrical Engineering and Informatics (ICAAEEI), 2024.
How to cite this paper
@article{1717878,
author = {Prince Vasoya, Dr. M. N. Nachappa},
title = {An Intelligent Air Quality Forecasting System Using Historical Data and Machine Learning},
journal = {Iconic Research And Engineering Journals},
year = {2026},
volume = {9},
number = {11},
pages = {2351-2359},
issn = {2456-8880},
url = {https://www.irejournals.com/formatedpaper/1717878.pdf},
abstract = {Air pollution is a significant environmental problem in urban areas, impacting millions of people's lives and leading to fatal and serious health effects. The World Health Organization (WHO) reports that almost 7 million people die from air pollution every year, and 90% of the world's population is breathing unhealthy air. While the Air Quality Index (AQI) helps represent air pollution levels in a standardised manner, most monitoring systems do not have any predictive ability for taking proactive decisions. This research introduces an intelligent air quality forecasting system with historical data and machine learning techniques to forecast the future air quality level. The system takes data from the major Indian cities from a multi-year period and incorporates certain pollutants (PM2.5, PM10, NO₂, SO₂, CO, O₃) and meteorological parameters. The data is preprocessed to deal with missing and inconsistent data. Various models such as regression models, ensemble models and deep learning models are tested. The results indicate that the advanced models are more accurate than the traditional ones, with XGBoost reaching an accuracy of more than 95% and the Bi-LSTM being able to capture temporal patterns well. The system provides 24-, 48-and 72-hour forecasts, which can be used for short term or medium-term planning. Particulate matter and meteorological factors are identified as prominent features in feature analysis. In general, it is scalable, and can help in making informed decisions for better air quality management and public health.},
keywords = {Air Quality Index (AQI), Machine Learning, Time Series Forecasting, XGBoost, LSTM, Particulate Matter (PM2.5/PM10), Air Pollution, Meteorological Parameters, Public Health, Ensemble Learning, India},
month = {May},
doi = {https://doi.org/10.64388/IREV9I11-1717878}
}