EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

A Novel TabPFN-Based Framework for Landslide Susceptibility Assessment: Multi-Model Comparison in Qingchuan County, Sichuan, China

Guogen Shui, Chong Li, Yixiang Du, Wenjun Zhang, Quanhong Zheng, Yitong Yao, Rong Hu, Yijun Lu, Jialun Cai

Applied Sciences · 06.08.2026 · 16 (15) · 7860

DOI: 10.3390/app16157860 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 06.08.2026. Verfasst von Guogen Shui und 8 weiteren Autor:innen. Die Arbeit nennt 37 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Technology Innovation Center for Emergency Surveying and Mapping, MNR, Central Government Guided Local Science and Technology Development Fund Projects, Youth Fund Project of Humanities and Social Sciences Research of the Ministry of Education of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Landslides are among the most common and destructive geological hazards in complex mountainous regions. Developing accurate and interpretable susceptibility assessment models is important for regional geological hazard prevention and spatial safety management. To address the limitations of traditional machine learning models in processing tabular geoscience data with limited samples and complex nonlinear relationships, this study took Qingchuan County, Sichuan Province, as the study area and introduced the Tabular Prior-data Fitted Network (TabPFN) for landslide susceptibility assessment. Its performance was systematically compared with four commonly used models: Logistic Regression (LR), Support Vector Machine (SVM), Random Forest (RF), and Extreme Gradient Boosting (XGBoost). An evaluation system was built using 12 conditioning factors, including elevation, lithology, and distance to roads (DTR). The SHapley Additive exPlanations (SHAP) method was further used to identify the contributions and effects of the main controlling factors. The results showed that TabPFN achieved the highest observed predictive performance among the evaluated models, with an AUC of 0.7834, showing higher observed performance than traditional models such as LR, SVM, RF, and XGBoost, indicating its strong ability to capture nonlinear relationships and good applicability under limited sample conditions. SHAP analysis further indicated that elevation, lithology, and distance to roads were the main controlling factors influencing the model predictions. This study demonstrates the applicability of TabPFN for landslide susceptibility assessment in complex mountainous regions and provides a new methodological reference for intelligent geological hazard prediction and refined regional risk management under limited data conditions.

Förderung

  • Technology Innovation Center for Emergency Surveying and Mapping, MNR
  • Central Government Guided Local Science and Technology Development Fund Projects
  • Youth Fund Project of Humanities and Social Sciences Research of the Ministry of Education of China

Zitieren

APA

Shui, G., Li, C., Du, Y., Zhang, W., Zheng, Q., Yao, Y., Hu, R., Lu, Y., & Cai, J. (2026). A Novel TabPFN-Based Framework for Landslide Susceptibility Assessment: Multi-Model Comparison in Qingchuan County, Sichuan, China. Applied Sciences, 16(15), 7860. https://doi.org/10.3390/app16157860

Vancouver

Shui G, Li C, Du Y, Zhang W, Zheng Q, Yao Y, et al. A Novel TabPFN-Based Framework for Landslide Susceptibility Assessment: Multi-Model Comparison in Qingchuan County, Sichuan, China. Applied Sciences. 2026;16(15):7860. doi:10.3390/app16157860

BibTeX
@article{Shui2026,
  author = {Shui, Guogen and Li, Chong and Du, Yixiang and Zhang, Wenjun and Zheng, Quanhong and Yao, Yitong and Hu, Rong and Lu, Yijun and Cai, Jialun},
  title = {A Novel TabPFN-Based Framework for Landslide Susceptibility Assessment: Multi-Model Comparison in Qingchuan County, Sichuan, China},
  journal = {Applied Sciences},
  year = {2026},
  volume = {16},
  number = {15},
  pages = {7860},
  doi = {10.3390/app16157860}
}
Kostenlos starten →