Predicting compressive strength of iron ore tailings concrete using data augmentation and machine learning
RONG Zhaochuan
XU Xiaochuan
LI Zhijun
DAI Jiahao
Abstract：This research examines the influence of Iron Ore Tailings(IOTs)on the compressive strength of concrete at 28 days.Employing a selection of four machine learning algorithms,a comprehensive analysis was undertaken.An initial selection of six pivotal variables facilitated the descriptive of 145 data sets,succeeded by data enhancement through the Variational Autoencoder(VAE).During the model development stage,a comparative assessment was made between Gradient Boosting,Adaptive Boosting,Random Forest,and Gaussian Process methodologies.Empirical findings underscored the Gradient Boosting algorithm's preeminence in performance metrics,notably in coefficient of determination and mean squared error criteria.The gradient boosting model-based prediction of the 28-day compressive strength of concrete has a high degree of consistency with actual data and strong generalization ability,provides scholars with an efficient,economical,and reliable prediction method.
Keywords：iron ore tailingscompressive strengthmachine learning techniquesdata enhancementgradient boosting methodology
Publication Date：2025-03-25
Online Publishing Date：2025-08-15(First online date of this platform, not the publication date of the document)
Pages：8( 95-101,141 )