Optimization-based multitarget stacked machine-learning model for estimating mechanical properties of conventional and fiber-reinforced preplaced aggregate concrete
Published: 14 May 2025| Version 1 | DOI: 10.17632/wt52w9g9p5.1
Contributors:
, Farzin Kazemi, , Description
The dataset employed in the study "Optimization-based multitarget stacked machine-learning model for estimating mechanical properties of conventional and fiber-reinforced preplaced aggregate concrete" consists of 475 data points, with 419 datasets allocated for PAC and 56 datasets designated for FR-PAC. These datasets have been meticulously gathered from both existing literature (389 and 42 for PAC and FR-PAC, respectively) and our experimental work with 30 datasets for PAC and 14 for FR-PAC. To use the dataset, you should refer to the original paper linked below to ensure licensing and credit.
Files
Institutions
University of Zintan, University of Tripoli, Military Technological College, Politechnika Gdanska, Jadara University
Categories
Data Science, Aluminum Silicate, Machine Learning, Machine Learning Algorithm, Concrete Technology, Reinforced Concrete, Steel-Reinforced Concrete, Fiber-Reinforced Composite, Slag, Fly Ash, Fiber Reinforced Concrete, Polypropylene, Infrastructure, Applied Machine Learning