Support Vector Machine (SVM) Application for Uniaxial Compression Strength (UCS) Prediction: A Case Study for Maragheh Limestone

The geomechanical properties of rock materials, such as uniaxial compression strength (UCS), are the main requirements for geo-engineering design and construction. A proper understanding UCS has a significant impression on safe different foundations rocks. So, applying fast reliable approaches to predict based limited data can be an efficient alternative regular traditional fitting curves. In order improve prediction accuracy UCS, presented study attempted utilize support vector machine (SVM) algorithm. Multiple training testing datasets were prepared predictions total 120 samples recorded limestone from Maragheh region, northwest Iran, which used achieve high precision rate prediction. models validated using confusion matrix, loss functions, error tables (MAE, MSE, RMSE). addition, 24 tested (20% primary dataset) model justifications. Referring results study, SVM (accuracy = 0.91/precision 0.86) showed good agreement with actual data, estimated coefficient determination (R2) reached 0.967, showing that model’s performance was impressively better than