Wildfire Occurrence Prediction Using Time Series Classification: A Comparative Study
Bibliographic record
Abstract
We compare the effectiveness of four machine learning models at predicting wildfire occurrence from multivariate time series containing hourly weather data, vegetation data, and fire occurrence data. Strategies to improve performance on highly imbalanced datasets are investigated, including adapting KNN and HMM to consider cost effectiveness. Two different training regimes are compared: the imbalanced training regime varies the class imbalance in the training and testing datasets together, and the balanced training regime keeps the imbalance ratio 50:50 for every training dataset. FCN and ResNet outperform KNN and HMM across all class imbalances tested. We tested the methods on the SaskFire dataset, which is an extensive, new dataset describing wildfires in Saskatchewan, Canada. The two models that performed best on highly imbalanced datasets are FCN and ResNet trained with the imbalanced training regime. On our dataset with a non-fire to fire class imbalance of 99:1, FCN and ResNet have precisions of 0.190 and 0.250, respectively, and recalls of 0.800.
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 | 0.008 | 0.013 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.005 | 0.003 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.002 | 0.003 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 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 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".