What does this research mean for the field?

A negative sampling method incorporating both credibility and spatial uniformity (CSUN) significantly improves the prediction accuracy and stability of machine learning models for landslide susceptibility assessment compared to random or credibility-only sampling. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

Develop and evaluate a negative sampling framework that combines credibility and spatial uniformity to improve machine learning-based landslide susceptibility assessment.

August 30, 2026Geomatics Natural Hazards and RiskOpen Access

A novel approach incorporating credibility and spatial uniformity for landslide negative sampling

View Full Paper

Ask AI

Bookmark

Authors

RWRuiting WangYunnan Normal University WXWenfei XiYunnan Normal University XGXiaojun Guo

Discussion

Member takes

Overview

Computational modeling study demonstrates improved predictive accuracy and stability for landslide susceptibility assessment, highlighting the value of balancing sample credibility with spatial...

Key Points

Develop and evaluate a negative sampling framework that combines credibility and spatial uniformity to improve machine learning-based landslide susceptibility assessment.
Segmented landslide feature spaces using K-means clustering and mapped environmental similarity to known landslides via a Gaussian similarity function.
Constructed a negative sample credibility map and selected negative samples ensuring both credibility and spatial uniformity (CSUN).
Trained Random Forest and Support Vector Machine models to compare the CSUN framework against random and credibility-only sampling through repeated sampling analysis.
CSUN improved the area under the receiver operating characteristic curve (AUC) by 20.5% compared to random sampling and by 8.7% compared to credibility-only sampling under the Random Forest model.
CSUN increased AUC by 12.3% versus random sampling and by 6.5% versus credibility-only sampling under the Support Vector Machine model.
Repeated sampling evaluations demonstrated that CSUN achieved the lowest standard deviations for both AUC and F1-score, establishing superior prediction stability.

Cite This Study

Wang et al. (2026) studied this question.

synapsesocial.com/papers/6a93f00a6c1a8fb52e79c17c https://doi.org/10.1080/19475705.2026.2721807

View Full Paper

Ask AI

Bookmark

August 30, 2026Geomatics Natural Hazards and RiskOpen Access

A novel approach incorporating credibility and spatial uniformity for landslide negative sampling

View Full Paper

Ask AI

Bookmark

Authors

RWRuiting WangYunnan Normal University WXWenfei XiYunnan Normal University XGXiaojun Guo

Discussion

Member takes

Overview

Computational modeling study demonstrates improved predictive accuracy and stability for landslide susceptibility assessment, highlighting the value of balancing sample credibility with spatial...

Key Points

Develop and evaluate a negative sampling framework that combines credibility and spatial uniformity to improve machine learning-based landslide susceptibility assessment.
Segmented landslide feature spaces using K-means clustering and mapped environmental similarity to known landslides via a Gaussian similarity function.
Constructed a negative sample credibility map and selected negative samples ensuring both credibility and spatial uniformity (CSUN).
Trained Random Forest and Support Vector Machine models to compare the CSUN framework against random and credibility-only sampling through repeated sampling analysis.
CSUN improved the area under the receiver operating characteristic curve (AUC) by 20.5% compared to random sampling and by 8.7% compared to credibility-only sampling under the Random Forest model.
CSUN increased AUC by 12.3% versus random sampling and by 6.5% versus credibility-only sampling under the Support Vector Machine model.
Repeated sampling evaluations demonstrated that CSUN achieved the lowest standard deviations for both AUC and F1-score, establishing superior prediction stability.

Cite This Study

Wang et al. (2026) studied this question.

synapsesocial.com/papers/6a93f00a6c1a8fb52e79c17c https://doi.org/10.1080/19475705.2026.2721807

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark