Machine Learning-Based Prediction of Compressive Strength of Cellulose Nanofiber-Reinforced Cement-Based Composites

This work presents a cross-validated machine learning framework for predicting the compressive strength of CNF-reinforced cement-based composites using a compiled experimental database of 695 samples, including 196 concrete, 266 cement paste, and 233 cement mortar observations. The input variables included cement, water, CNFs, superplasticizer, fine aggregate, coarse aggregate, and curing age, depending on the material type. Ten supervised regression models were evaluated: random forest, linear regression, support vector regression, gradient boosting regressor, AdaBoost regressor, k-nearest neighbors regressor, bagging regressor, extreme gradient boosting regressor, decision tree, and pruned decision tree.