Three machine learning algorithms,namely BP neural network model,support vector machine,and XGBoost model were selected to construct a prediction model between the amount of concrete raw materials and the 28 d compressive strength.The fitting degree,generalization ability,and prediction error of each model were compared.The results show that the XGBoost model performs well in terms of fitting and generalization ability,and it is the best model for predicting the relationship between the amount of concrete raw materials and the 28 d compressive strength.
维普
万方数据