A comparative investigation using artificial neural network (ANN) and decision tree (DT) methods in the prediction of slump and strength for concrete samples

Abstract

In the past few years, the application of Machine Learning Techniques (MLT) has become a popular way to enhance the accuracy of predicting concrete properties. This study aims to compare and contrast the performance of Artificial neural network (ANN) and Decision Tree (DT) methods in predicting the compressive strength and slump values of concrete samples. Experimental data used for model building and comparison were obtained from a previous research project. R-squared value (RSQ) and Mean Squared Error (MSE) metrics were used to determine which regression method was the most efficient in predicting concrete compressive strength and slump values. The results from the comparison between ANN and DT methods would be able to identify which of the two regression models is the better choice for forecasting concrete properties.