Predicting Angle of Internal Friction and Cohesion of Rocks Based on Machine Learning Algorithms

The safe and sustainable design of rock slopes, open-pit mines, tunnels, foundations, underground excavations requires appropriate reliable estimation strength deformation characteristics. Cohesion (????) angle internal friction (????) are the two key parameters widely used to characterize shear materials. Thus, prediction these is essential evaluate stability any formation. In this study, four advanced machine learning (ML)-based intelligent models, namely Lasso regression (LR), ridge (RR), decision tree (DT), support vector (SVM), were developed predict ???? in (MPa) ???? (°), with P-wave velocity (m/s), density (gm/cc), UCS (MPa), tensile as input parameters. actual dataset having 199 data points no missing was allocated identically for each model 70% training 30% testing purposes. To enhance performance an iterative 5-fold cross-validation method used. coefficient determination (R2), mean absolute error (MAE), square (MSE), root (RMSE), a10-index metrics optimal model. results revealed SVM be a more efficient predicting (R2 = 0.977) 0.916) than LR (????: R2 0.928 ????: 0.606), RR 0.961 0.822), DT 0.934 0.607) on data. Furthermore, check level accuracy model, sensitivity analysis performed showed that most influential ????. findings study contribute long-term evaluation masses surface subsurface excavations.