.png){kind=icon}
.png){kind=icon}
Return
Deep forest-based hybrid model for tunnel squeezing risk prediction using ensemble learning
.png){kind=icon}
PRE
AI
J
D
P
Z
X
F
B
DOI：10.1016/j.autcon.2026.107191.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A hybrid Deep Forest model achieves 95.24% accuracy in tunnel squeezing prediction. • Borderline-SMOTE and IARO enhance model performance under data imbalance. • SHAP–ICE analysis reveals nonlinear impacts of SSR and rock mass quality. • A user-friendly GUI enables real-time and explainable tunnel risk assessment.
Keywords:
Tunnel squeezing prediction
Deep forest
Borderline-SMOTE
Metaheuristic algorithm
Bayesian optimization
Journal
IF:
11.5
Papers:
6.1K
Citations:
4.2W
Organization
Cited Papers