Experimental Dataset for Predicting the Bearing Capacity of Rubber Fiber-Reinforced Ring Foundations Using Ensemble Machine Learning

This dataset contains the experimental database used for developing and validating an interpretable ensemble machine learning framework for predicting the bearing capacity of rubber fiber-reinforced ring foundations. The dataset includes 128 experimental records obtained from laboratory investigations of circular and ring foundations under different geometric configurations and reinforcement conditions. The input parameters include radius ratio, rubber fiber content, reinforced layer thickness, soil cohesion, friction angle, and soil density. The output variable is the ultimate bearing capacity of the foundation. The dataset was used for machine learning model development, performance evaluation, and SHAP-based explainable analysis in the associated research article. The dataset is provided to support reproducibility and facilitate further research on data-driven prediction of geotechnical foundation behavior.