GUI-powered compressive strength estimation of green concrete utilising an efficient ensemble learning paradigm

This study develops an efficient ensemble learning paradigm (ELP)-based graphical user interface (GUI) framework to estimate the compressive strength (CS) of green concrete. In this study, eight ELPs were employed and the best-fitted model was chosen to estimate the CS of green concrete. Initially, Adaboost regressor, decision tree regressor, gradient boosting regressor, random forest regressor, and support vector regressor were employed. In addition, a bagging-based approach was used to construct four additional ELPs. Experimental results show that the employed GBR achieved the most accurate precision in both training (R = 0.9941) and testing (R = 0.9669) phases. The outcomes of the GBR paradigm are significantly superior to those of the baseline model, i.e. linear regressor and support vector regressor. The overall results demonstrate that the utilised GBR paradigm can potentially estimate the CS of fly ash and slag-based green concrete with a high degree of precision and robustness. The developed GUI (see supplementary material) provides an efficient tool for estimating the CS of green concrete across single, multiple, and predefined datasets. Researchers and practitioners can use the developed GUI to estimate the concrete CS based on pre-defined values of design mix parameters such as the contents of cement, aggregated and FAsh.