• Home
  • Search
  • Improving the Generalization Performance of Debris-Flow Susceptibility Modeling by a Stacking Ensemble Learning-Based Negative Sample Strategy
  • Cite Icon1
  • https://doi.org/10.3390/w17162460Copy DOI Icon

Improving the Generalization Performance of Debris-Flow Susceptibility Modeling by a Stacking Ensemble Learning-Based Negative Sample Strategy

  • Aug 19, 2025
  • Water
  • Jiayi Li +4 more
Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

To address the negative sample selection bias and limited interpretability of traditional debris-flow event susceptibility models, this study proposes a framework that enhances generalization by integrating negative sample screening via a stacking ensemble model with an interpretable random forest. Using Wenchuan County, Sichuan Province, as the study area, 19 influencing factors were selected, encompassing topographic, geological, environmental, and anthropogenic variables. First, a stacking ensemble—comprising logistic regression (LR), decision tree (DT), gradient boosting decision tree (GBDT), and random forest (RF)—was employed as a preliminary classifier to identify very low-susceptibility areas as reliable negative samples, achieving a balanced 1:1 ratio of positive to negative instances. Subsequently, a stacking–random forest model (Stacking-RF) was trained for susceptibility zonation, and SHAP (Shapley additive explanations) was applied to quantify each factor’s contribution. The results show that: (1) the stacking ensemble achieved a test-set AUC (area under the receiver operating characteristic curve) of 0.9044, confirming its effectiveness in screening dependable negative samples; (2) the random forest model attained a test-set AUC of 0.9931, with very high-susceptibility zones—covering 15.86% of the study area—encompassing 92.3% of historical debris-flow events; (3) SHAP analysis identified the distance to a road and point-of-interest (POI) kernel density as the primary drivers of debris-flow susceptibility. The method quantified nonlinear impact thresholds, revealing significant susceptibility increases when road distance was less than 500 m or POI kernel density ranged between 50 and 200 units/km2; and (4) cross-regional validation in Qingchuan County demonstrated that the proposed model improved the capture rate for high/very high susceptibility areas by 48.86%, improving it from 4.55% to 53.41%, with a site density of 0.0469 events/km2 in very high-susceptibility zones. Overall, this framework offers a high-precision and interpretable debris-flow risk management tool, highlights the substantial influence of anthropogenic factors such as roads and land development, and introduces a “negative-sample screening with cross-regional generalization” strategy to support land-use planning and disaster prevention in mountainous regions.

Similar Papers
  • Research Article
  • Citations1

A machine learning model to predict treatment initiation among new patients in a community oncology network.

  • Jun 01, 2023
  • Journal of Clinical Oncology
  • Bo He +7
  • Research Article
  • Citations5

Dynamic Strength Prediction of Brittle Engineering Materials via Stacked Multi-Model Ensemble Learning and Interpretability-Driven Feature Analysis.

  • Jun 27, 2025
  • Materials (Basel, Switzerland)
  • Xin Cai +6
  • Research Article

Predicting overall survival in synchronous metastatic nasopharyngeal carcinoma using a stacking ensemble machine learning model: a multicenter retrospective study.

  • Jan 01, 2026
  • Frontiers in oncology
  • Canwen Che +9
  • Research Article

Research on Advertising Prediction Based on Stacking Ensemble Algorithms

  • Jan 15, 2025
  • Theoretical and Natural Science
  • Sirui Fang
  • Research Article
  • Citations4

Predictive value of the stone-free rate after percutaneous nephrolithotomy based on multiple machine learning models

  • Aug 19, 2025
  • Frontiers in Medicine
  • Zhao Rong Liu +3
  • PDF
  • Research Article
  • Citations4

Predicting potential miRNA-disease associations based on more reliable negative sample selection

  • Oct 17, 2022
  • BMC Bioinformatics
  • Ruiyu Guo +4
  • Research Article
  • Citations2

A stacking ensemble model based on Elastic Net and Random Forest for box office prediction

  • Jul 05, 2024
  • Applied and Computational Engineering
  • Weichen Fu
  • Research Article

A hybrid machine learning approach for reliably predicting surface roughness in CNC turning operations

  • Mar 06, 2026
  • Scientific Reports
  • Hakan Yurtkuran +4
  • Abstract

P-027: Development of an artificial intelligence diagnostic model based on routine laboratory results and echocardiography for the early diagnosis of light chain amyloidosis

  • Oct 01, 2021
  • Clinical Lymphoma Myeloma and Leukemia
  • Wei Yang +6
  • Research Article
  • Citations56

Stacking ensemble of machine learning methods for landslide susceptibility mapping in Zhangjiajie City, Hunan Province, China

  • Dec 28, 2022
  • Environmental Earth Sciences
  • Yuke Huan +3
  • Research Article
  • Citations16

Prediction of gully erosion susceptibility through the lens of the SHapley Additive exPlanations (SHAP) method using a stacking ensemble model.

  • May 01, 2025
  • Journal of environmental management
  • Jeongho Han +2
  • Research Article

An Interpretable Stacking Ensemble Model with SHAP for Geriatric Depression Prediction: Analysis of the NHANES 2005-2023 Database<b></b>

  • Jan 28, 2026
  • EAI Endorsed Transactions on Pervasive Health and Technology
  • Manman Cui +2
  • Preprint Article

Machine Learning-Based Predictive Models for Identifying Fetal Growth Restriction in Patients With Early-Onset Preeclampsia: Retrospective Study (Preprint)

  • Jul 17, 2025
  • Ying Zhang +7
  • Research Article
  • Citations1

Prediction of 180-day post-admission’s GOS in patients with spontaneous intracerebral hemorrhage based on explainable machine learning technology

  • Nov 21, 2025
  • Medicine
  • Cuijie Yuan +8
  • Research Article
  • Citations4

An explainable and supervised machine learning model for prediction of red blood cell transfusion in patients during hip fracture surgery

  • Dec 19, 2024
  • BMC Anesthesiology
  • Yongchang Zhou +7
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.