Performance comparison of post-earthquake disaster susceptibility assessment models based on GIS: a case study of the Lushan County in Ya’an City, China
Bibliographic record
Abstract
To verify the rationality of different mathematical models for susceptibility assessment of post-earthquake disasters, taking the hazard statistics of Lushan County in Ya'an City, China as an example, this study preliminarily selected 12 evaluation factors closely related to geological disasters, such as elevation, slope, and aspect, as well as 159 actual disaster sites based on Geographic Information System (GIS). Traditional susceptibility assessment models-including the information value (I) model, certain factor (CF) model, informative-logistic regression (I-LR) model, and certain factor-logistic regression (CF-LR) model-were applied, alongside the machine learning-based random forest (RF) model, to evaluate the susceptibility of local disasters. Thirty disaster sites that were not included in the models were selected as test samples, and the rationality and accuracy of the four mathematical models were evaluated and tested using the frequency ratio method and Receiver Operating Characteristic (ROC) curve method, respectively. The results showed that all four traditional models indicated that the extremely high and high susceptibility areas of Lushan County are mainly concentrated in the low-altitude and valley areas in the south and central parts, while the low and extremely low susceptibility areas are distributed in the high mountainous and canyon areas in the north, which is basically consistent with the actual investigation. The AUC values for the four traditional models' evaluation accuracy, from high to low, are CF-LR (0.825), I-LR (0.822), I (0.816), and CF (0.815). The first two coupled models, which consider the weight coefficients of influencing factors, show slightly improved results compared to single models. However, the Random Forest model based on machine learning has an AUC value of 0.920 for evaluation accuracy, demonstrating the best performance. The research findings offer valuable insights for selecting regional susceptibility assessment models for post-earthquake disasters.
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 | 0.004 | 0.007 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.003 | 0.001 |
| Science and technology studies | 0.000 | 0.001 |
| Scholarly communication | 0.002 | 0.002 |
| 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".