EUVIMEDSkopri EUVIMED
Video/Audio

Hybrid Deep Learning with Attention-Based Interpretability for PM2.5 Forecasting in Delhi Urban Environments

Mahasiva Saravagna Sai Lakshmi, Gollapudi Svssvn Rithvik, Ch V V Ramana, N Thirupathi Rao, Faten Khalid Karim, Ayman Aljarbouh, Samih M Mostafa

Journal of Visualized Experiments · 2026

DOI: 10.3791/71004 ↗

At a glance

Video/Audio in Journal of Visualized Experiments, 2026. Written by Mahasiva Saravagna Sai Lakshmi and 6 co-authors. Institution: Department of AI&DS, Vignan's Institute of Information Technology. Major MEDLINE topics: Particulate Matter, Deep Learning, Environmental Monitoring, Air Pollutants.

AstrattSubjectsDettalji biblijografiċiCiteRelated studies

Abstract (original language)

Short-term PM2.5 forecasting is challenging in Delhi because particulate matter concentrations are influenced by local emissions, meteorological variation, seasonal stagnation, and episodic fire-related pollution. This study presents CORTA-Net, a hybrid deep learning framework for PM2.5 forecasting using multi-source environmental data from 2012 to 2024. The input data includes hourly air-quality observations from CPCB (Central Pollution Control Board) /DPCC (Delhi Pollution Control Committee) monitoring stations, meteorological variables from the India Meteorological Department (IMD), and satellite-derived fire-count information from MODIS (Moderate Resolution Imaging Spectroradiometer) products. The proposed framework first applies CorrXGBoost-Rank feature selection to reduce redundant predictors and retain important pollutants, meteorological, temporal, and fire-related variables. The selected features are then arranged as supervised sliding-window sequences and processed using a transfer-learning-based LSTM encoder followed by a multi-head attention layer. The attention mechanism provides feature- and time-step-level interpretation of the PM2.5 forecast. CORTA-Net was evaluated using chronological training, testing, and validation partitions, as well as cross-validation. Compared with random forest, XGBoost, LSTM, and attention-LSTM baselines, the proposed framework reduced prediction error under the evaluated Delhi monitoring-station setting. The novelty of CORTA-Net lies in combining explicit CorrXGBoost-Rank feature screening, transfer-learning-based temporal encoding, MODIS fire-activity integration, and multi-head attention-based model-behavior analysis in a single reproducible PM2.5 forecasting pipeline. Practically, the framework can support short-term urban air-quality forecasting in data-rich monitoring environments where pollutant records, meteorological observations, and fire-activity indicators are available.

Subjects

MeSH terms (MEDLINE)

Particulate MatterDeep LearningEnvironmental MonitoringAir PollutantsIndiaForecastingCitiesAir Pollution

Substances

Particulate MatterAir Pollutants

Authors and institutions

  • Department of AI&DS, Vignan's Institute of Information Technology.
  • Vignan's Institute of Information Technology; challa@vignaniit.edu.in.
  • Department of Computer Science and Engineering, Vignan's Institute of Information Technology.
  • Department of Computer Sciences, College of Computer and Information Sciences, Princess Nourah bint Abdulrahman University.
  • Department of Software Engineering, College of Computer Engineering and Science, Prince Mohammad Bin Fahd University.
  • Computer Science Department, Faculty of Computers and Information, Qena University.

Cite

APA

Saravagna Sai Lakshmi, M., Svssvn Rithvik, G., Ramana, C. V. V., Rao, N. T., Karim, F. K., Aljarbouh, A., & Mostafa, S. M. (2026). Hybrid Deep Learning with Attention-Based Interpretability for PM2.5 Forecasting in Delhi Urban Environments. Journal of Visualized Experiments. https://doi.org/10.3791/71004

Vancouver

Saravagna Sai Lakshmi M, Svssvn Rithvik G, Ramana CVV, Rao NT, Karim FK, Aljarbouh A, et al. Hybrid Deep Learning with Attention-Based Interpretability for PM2.5 Forecasting in Delhi Urban Environments. Journal of Visualized Experiments. 2026. doi:10.3791/71004

BibTeX
@article{SaravagnaSaiLakshmi2026,
  author = {Saravagna Sai Lakshmi, Mahasiva and Svssvn Rithvik, Gollapudi and Ramana, Ch V V and Rao, N Thirupathi and Karim, Faten Khalid and Aljarbouh, Ayman and Mostafa, Samih M},
  title = {Hybrid Deep Learning with Attention-Based Interpretability for PM2.5 Forecasting in Delhi Urban Environments},
  journal = {Journal of Visualized Experiments},
  year = {2026},
  doi = {10.3791/71004}
}
Start for free →