Abstract
The use of rice husk ash(RHA)in concrete serves a positive role. The compressive strength of RHA in concrete is predicted using supervised machine learning approaches such as decision trees (DT), bagging regressors, and AdaBoost regressors in this research. For the evaluation of the performance of developed models, following checks such as R2, mean absolute error (MAE), root mean square error (RMSE), and root mean square log error (RMSLE) were calculated. To further verify the model's accuracy, k-fold cross-validation is used. Comparing to the DT and Adaboostmodel, the bagging regressor model is much more successful in forecasting theresults having R2 value of 0.93. The error values that were found for MAE, RMSE, RMSLE were lesser, while greater values of the R2 were the unambiguous signs of the superior performance of the model. Furthermore, the impact of each six input parameter on the outcome was calculated using the sensitivity analysis approach. Using machine learning will save time, labor, and resources in civil engineering for predicting mechanical properties of concrete.