Explainable Ensemble Learning Approaches for Predicting the Compression Index of Clays

Abstract

Accurate prediction of the compression index (cc) is essential for geotechnical infrastructure design, especially in clay-rich coastal regions. Traditional methods for determining cc are often time-consuming and inconsistent due to regional variability. This study presents an explainable ensemble learning framework for predicting the cc of clays. Using a comprehensive dataset of 1080 global samples, four key geotechnical input variables—liquid limit (LL), plasticity index (PI), initial void ratio (e0), and natural water content w—were leveraged for accurate cc prediction. Missing data were addressed with K-Nearest Neighbors (KNN) imputation, effectively filling data gaps while preserving the dataset’s distribution characteristics. Ensemble learning techniques, including Random Forest (RF), Gradient Boosting Decision Trees (GBDT), Extreme Gradient Boosting (XGBoost), and a Stacking model, were applied. Among these, the Stacking model demonstrated the highest predictive performance with a Root Mean Squared Error (RMSE) of 0.061, a Mean Absolute Error (MAE) of 0.043, and a Coefficient of Determination (R2) value of 0.848 on the test set. Model interpretability was ensured through SHapley Additive exPlanations (SHAP), with e0 identified as the most influential predictor. The proposed framework significantly improves both prediction accuracy and interpretability, offering a valuable tool to enhance geotechnical design efficiency in coastal and clay-rich environments.