Data Augmentation Strategies for Improved PM<sub>2.5</sub> Forecasting Using Transformer Architectures
Bibliographic record
Abstract
Breathing in fine particulate matter with diameters less than 2.5 µm (PM2.5) has greatly increased an individual’s risk of cardiovascular and respiratory diseases. As climate change progresses, extreme weather events, including wildfires, are expected to rise, exacerbating air pollution. The 2023 Canadian wildfires highlighted the growing threat of PM2.5 as smoke spread across U.S. cities like New York, Philadelphia, and Washington D.C. This research investigates the application of data augmentation techniques to improve the accuracy of PM2.5 concentration forecasts in these urban environments. Models trained on imbalanced datasets often struggle to capture extreme pollution events, underestimating high PM2.5 levels due to the model’s focus on more frequent, low-value samples. To address this, we implemented cluster-based undersampling and trained transformer models using various cutoff thresholds (12.1 µg/m³ and 35.5 µg/m³) and partial sampling ratios (10/90, 20/80, 30/70, 40/60, 50/50). Our results demonstrate that the 35.5 µg/m³ threshold, coupled with a 20/80 partial sampling ratio, provides the best performance regarding RMSE and R², particularly in capturing high PM2.5 events. Overall, models trained on augmented data significantly outperformed those trained on original data, highlighting the importance of resampling techniques in improving air quality forecasting accuracy, especially for high-pollution scenarios. These insights significantly contribute to a better understanding of PM2.5 pollution with the hopes of more informed public health and environmental policies.
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 machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.002 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.002 | 0.001 |
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 source (direct Gemma or distilled Codex), 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".