Spatio-Temporal Agnostic Sampling for Imbalanced Multivariate Seasonal Time Series Data: A Study on Forest Fires
Bibliographic record
Abstract
Natural disasters are mostly seasonal and caused by anthropological, climatic, and geological factors that impact human life, economy, ecology, and natural resources. This paper focuses on increasingly widespread forest fires, causing greater destruction in recent years. Data obtained from sensors for predicting forest fires and assessing fire severity, i.e., area burned, are multivariate, seasonal, and highly imbalanced with a ratio of 100,000+ non-fire events to 1 fire event. This paper presents Spatio-Temporal Agnostic Sampling (STAS) to overcome the challenge of highly imbalanced data. This paper first presents a mathematical understanding of fire and non-fire events and then a thorough complexity analysis of the proposed STAS framework and two existing methods, NearMiss and SMOTE. Further, to investigate the applicability of STAS, binary classification models (to determine the probability of forest fire) and regression models (to assess the severity of forest fire) were built on the data generated from STAS. A total of 432 experiments were conducted to validate the robustness of the STAS parameters. Additional experiments with a temporal data split were conducted to further validate the results. The results show that 180 of the 216 binary classification models had an F1score>0.9 and 150 of the 216 regression models had an R2score>0.75. These results indicate the applicability of STAS for fire prediction with highly imbalanced multivariate seasonal time series data.
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 | 0.005 | 0.009 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.000 | 0.001 |
| 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 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".