A Hybrid IEFDL Model for Accurate PM2.5 Forecasting in India Using Deep Learning and Multivariate Data Analysis

  • Authors

    • Vaddadi Vasudha Rani Department of Information Technology, GMR Institute of Technology(A), Rajam, Andhra Pradesh, India
    • M. Seshashaye Department of Computer Science, GSS, GITAM (Deemed to be University), Visakhapatnam, ‎ Andhra Pradesh, India
    • Yarlagadda Anuradha Department of Computer Science and Engineering, Gayatri Vidya Parishad College of Engineering (A), Visakhapatnam, ‎Andhra Pradesh, India
    • Madhavi Kolukuluri Department of Computer Science and Engineering, Nadimpalli Satyanarayana Raju Institute of Technology (A), Visakhapatnam, Andhra Pradesh, India
    • Gudepu Sridevi Department of Mechanical Engineering, Vignan’s Institute of Information Technology (A), Visakhapatnam, Andhra Pradesh, India
    • Chinta Manjusha Department of Electronics and Communication Engineering, Koneru Lakshmaiah Education Foundation, KL (Deemed to be University), Vaddeswaram, Guntur District, Andhra Pradesh, India
    • Chandra Sekhar Musinana Department of Information Technology, MVGR College of Engineering, Vizianagaram, Andhra Pradesh, India
    https://doi.org/10.14419/cvc00e96

    Received date: May 28, 2025

    Accepted date: June 20, 2025

    Published date: June 23, 2025

  • Particulate Matter (PM2.5), Improved Extraction of Feature with Deep Learning, Recurrent Neural Network (RNN), XGBoost, Dimensionality Reduction (PCA), Air Quality Prediction

  • Abstract

    Air pollution is rising rapidly due to rapid industrialization, making PM forecasting, especially PM2.5, a key research subject due to its health implications. An Improved Extraction of Feature with a Deep Learning model, also named as the IFEDL model, is presented for Indian city particulate matter forecasting. The model combines historical air quality data with environmental and socioeconomic data using hybrid deep learning. Deep learning models like RNNs for sequential historical data and DBNs for multidimensional factor analysis are used for long-term projections. Short-term forecasts employ XGBoost because it works well with limited data and characteristics. Reduced dimension improves model performance with PCA. In addition to weather, industrial expansion, car sales, and emissions, the National Ambient Air Quality database offers hourly PM data for many years. Train and test the IEFDL model 80-20. Compared to standard models, experimental results revealed improved short-term prediction accuracy with a Mean Absolute Error of 4.92 µg/m³ and Root Mean Square Error of 6.75 µg/m³. Combining several factors and deep learning architectures yielded accurate long-term projections, with an MAE of 6.34 µg/m³ and RMSE of 8.89 µg/m³. In rising nations like India, the concept promotes proactive air quality control and policymaking.

  • References

    1. M. Catalano, F. Galatioto, M. Bell, A. Namdeo, and A. S. Bergantino, “Improving the prediction of air pollution peak episodes generated by urban transport networks,” Environmental Science & Policy, vol. 60, pp. 69–83, 2016, https://doi: 10.1016/j.envsci.2016.03.008.
    2. Q. Tao, F. Liu, Y. Li, and D. Sidorov, “Air pollution forecasting using a deep learning model based on 1D Convnets and bidirectional GRU,” IEEE Access, vol. 7, pp. 76690–76698, 2019, https://doi: 10.1109/access.. 2019.2921578.
    3. L. Bai, J. Wang, X. Ma, and H. Lu, “Air Pollution Forecasts: An Overview,” International Journal of Environmental Research and Public Health, vol. 15, no. 4, p. 780, 2018, https://doi: 10.3390/ijerph15040780.
    4. T. Xayasouk, H. Lee, and G. Lee, “Air pollution prediction using Long Short-Term Memory (LSTM) and Deep Autoencoder (DAE) models,” Sus-tainability, vol. 12, no. 6, p. 2570, 2020, https://doi: 10.3390/su12062570.
    5. Y.-T. Tsai, Y.-R. Zeng, and Y.-S. Chang, “Air Pollution Forecasting Using RNN with LSTM,” 2018 IEEE 16th Intl Conf on Dependable, Auto-nomic and Secure Computing, 16th Intl Conf on Pervasive Intelligence and Computing, 4th Intl Conf on Big Data Intelligence and Computing and Cyber Science and Technology Congress, pp. 1074–1079, Aug. 2018, https://doi: 10.1109/dasc/picom/datacom/cyberscitec.2018.00178.
    6. P. Jiang, C. Li, R. Li, and H. Yang, “An innovative hybrid air pollution early-warning system based on pollutants forecasting and Extenics evalua-tion,” Knowledge-Based Systems, vol. 164, pp. 174–192, 2019, https://doi: 10.1016/j.knosys.2018.10. 036..
    7. R. Dong, R. Fisman, Y. Wang, and N. Xu, “Air pollution, affect, and forecasting bias: Evidence from Chinese financial analysts,” Journal of Finan-cial Economics, vol. 139, no. 3, pp. 971–984, 2019, https://doi: 10.1016/j.jfineco.2019.12.004.
    8. R. Rakholia, Q. Le, K. Vu, B. Q. Ho, and R. S. Carbajo, “Accurate PM2.5 urban air pollution forecasting using multivariate ensemble learning Ac-counting for evolving target distributions,” Chemosphere, vol. 364, p. 143097, 2024, https://doi: 10.1016/j.chemosphere.2024.143097.
    9. K. Damkliang and J. Chumnaul, “Deep learning and statistical approaches for area-based PM2.5 forecasting in Hat Yai, Thailand,” Journal of Big Data, vol. 12, no. 1, 2025, https://doi: 10.1186/s40537-025-01079-9.
    10. N. Zaini, L. W. Ean, A. N. Ahmed, M. A. Malek, and M. F. Chow, “PM2.5 forecasting for an urban area based on deep learning and decomposi-tion method,” Scientific Reports, vol. 12, no. 1, 2022, https://doi: 10.1038/s41598-022-21769-1.
    11. F. Mohammadi, H. Teiri, Y. Hajizadeh, A. Abdolahnejad, and A. Ebrahimi, “Prediction of atmospheric PM2.5 level by machine learning techniques in Isfahan, Iran,” Scientific Reports, vol. 14, no. 1, 2024, https://doi: 10.1038/s41598-024-52617-z.
    12. I. Yeo, Y. Choi, Y. Lops, and A. Sayeed, “Efficient PM2.5 forecasting using geographical correlation based on integrated deep learning algorithms,” Neural Computing and Applications, vol. 33, no. 22, pp. 15073–15089, 2021, https://doi: 10.1007/s00521-021-06082-8.
    13. K. Sravanthi, V. Ch, L. K. Kumar, B. Keerthana, B. Munukurthi, and B. Samatha, “A Novel Approach to Enhancing Air Pollution Prediction using a Two-Stage Neural XG Boost Detection Algorithm,” International Journal of Basic and Applied Sciences, vol. 14, no. 1, pp. 99–105, 2025, doi: 10.14419/a2299s64.
    14. Z. Zhang and S. Zhang, “Modeling air quality PM2.5 forecasting using deep sparse attention-based transformer networks,” International Journal of Environmental Science and Technology, vol. 20, no. 12, pp. 13535–13550, Apr. 2023, https://doi: 10.1007/s13762-023-04900-1.
    15. N. Doreswamy, H. K. S, Y. Km, and I. Gad, “Forecasting air pollution particulate matter (PM2.5) using machine learning regression models,” Pro-cedia Computer Science, vol. 171, pp. 2057–2066, Jan. 2020, doi: 10.1016/j.procs.2020.04.221.
    16. M. S. Rao, B. Sailaja, M. Swetha, G. Kumari, B. Keerthana, and B. Sambana, “Statistical Approaches for Forecasting Air pollution: A Review,” Springer Proceedings in Mathematics & Statistics, pp. 37–44, Jan. 2024, https://doi.org/10.1007/978-3-031-51163-9_5.
    17. Z. Guo, C. Yang, D. Wang, and H. Liu, “A novel deep learning model integrating CNN and GRU to predict particulate matter concentrations,” Process Safety and Environmental Protection, vol. 173, pp. 604–613, Mar. 2023, https://doi: 10.1016/j.psep.2023.03.052.
    18. V. Singh, S. K. Sahana, and V. Bhattacharjee, “A novel CNN-GRU-LSTM based deep learning model for accurate traffic prediction,” Deleted Journal, vol. 28, no. 1, Apr. 2025, https://doi: 10.1007/s10791-025-09526-0.

  • Downloads

  • How to Cite

    Rani , V. V. ., Seshashaye , M. ., Anuradha , Y. ., Kolukuluri , M. ., Sridevi , G. ., Manjusha , C. ., & Musinana , C. S. . (2025). A Hybrid IEFDL Model for Accurate PM2.5 Forecasting in India Using Deep Learning and Multivariate Data Analysis. International Journal of Basic and Applied Sciences, 14(2), 344-351. https://doi.org/10.14419/cvc00e96