Gradient boosting-based regression modelling for estimating the time period of the irregular precast concrete structural system with cross bracing

Abstract

Regression tree, random forest, bagging and gradient boosting regression-based modelling techniques were used to estimate the time period of precast concrete frame structures with vertical irregularity and cross-bracing using 756 Etabs models. This paper thoroughly explored the efficacy of random forest, regression tree, bagging, gradient boosting-based modelling using RStudio. The time period has been the output parameter while the number of cross-bracing, column size, beam size, soft storey, irregularity coefficient and height of the building has been assigned as input parameters. The accuracy of machine learning techniques has been checked by reference to the formulas published in the literature. A comparison of results indicates that the gradient boosting-based regression approach performed well as compared to random forest regression, bagging and regression tree.