Air Pollution Forecasting Using Machine Learning with Temporal Fusion Transformer and Graph Neural Networks
Bibliographic record
Abstract
Air pollution is a significant obstacle to achieving urban sustainability and public health, particularly in rapidly developing regions. Accurate air quality prediction is essential for proactive pollution management and policy formulation. In this study, we employ sophisticated machine learning frameworks, such as Temporal Fusion Transformer (TFT) and Graph Neural Networks (GNNs), to predict air quality levels. TFT is designed to capture temporal dependencies, making it ideal for time-series forecasting, while GNNs model spatial relationships, enabling the analysis of pollutant dispersion across monitoring stations. A synthetic dataset comprising 50 rows was generated, simulating pollutant concentrations (PM2.5, PM10, CO, NO2, SO2) and meteorological variables (temperature, humidity, wind speed). The models were assessed using criteria like <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$R^{2}$</tex>, RMSE, and MAE. TFT achieved an <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$R^{2}$</tex> of 0.97 and an RMSE of <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$4.2 \mu ~\mathrm{g} / \mathrm{m}^3$</tex>, significantly outperforming traditional models like Random Forest and XGBoost. GNN provided spatial insights, identifying pollutant hotspots and their dispersion patterns. This research highlights the superior accuracy and scalability of TFT and GNN for air quality prediction. The results underline their potential for real-time monitoring, smart city integration, and proactive environmental management, offering a robust solution for addressing urban air pollution challenges.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".