Unveiling global flood hotspots: Optimized machine learning techniques for enhanced flood susceptibility modeling
Bibliographic record
Abstract
Worldwide Floods are among the most catastrophic and dangerous natural calamities globally, causing irreparable damage to human lives and property, and environmental degradation. Flood susceptibility mapping is a cost-effective tool to mitigate and manage the impacts of flood occurrences, but high accuracy in mapping is important to support management strategies. This study assessed the efficiency of three machine learning approaches, including support vector regression (SVR) and its optimized versions through combination with grey wolf optimizer (GWO) and whale optimization algorithm (WOA), in generating accurate flood susceptibility maps at a global scale. Data from 6682 historical flood events, covering eight flood-related geo-environmental factors were used to generate the maps. All maps produced were evaluated based on root mean square error (RMSE), mean squared error (MSE), standard deviation, and area under the receiver operating characteristic curve (AUC). This study reveals that the SVR-GWO model has the best performance in predicting flood-prone areas worldwide based on AUC , RMSE and MSE. The findings indicate that approximately 17.14 % of global land area is highly and very highly susceptible to flood occurrence. Flood hot-spot countries were the United States of America (7.75 %), Indonesia (6.33 %), India (6.31 %), Brazil (5.33 %) and Nigeria (4.08 %). Countries with the lowest probability of flood occurrence were the Russian Federation, Canada, Greenland, the United States of America and China. Incorporating additional satellite-based environmental data could further enhance the model's accuracy. Furthermore, the approach sets a foundation for future research in tailoring flood prediction models to regional scales, addressing the diverse challenges posed by different geographic and environmental settings. • SVR-GWO model outperformed others in predicting global flood-susceptible areas. • United States of America, Indonesia, and India were top flood-prone countries. • Distance to river, land use, and rainfall were key flood susceptibility factors.
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 | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 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".