Prediction model for compressive strength of alkali-activated multi-source solid-waste grouting materials based on generative adversarial networks and machine learning

Qiang Yue, Guangming Yu, Yong Yao, Zhiqi Yang, Yufeng Li, Liyang Bai, Guangwei Xiong · 2026

Read Prediction model for compressive strength of alkali-activated multi-source solid-waste grouting materials based on generative adversarial networks and machine learning by Qiang Yue, Guangming Yu, Yong Yao, Zhiqi Yang, Yufeng Li, Liyang Bai, Guangwei Xiong (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0357051.txt.

Description

Qiang Yue and coauthors published a 2026 materials paper in PLOS ONE on predicting the strength of grouting materials made from industrial solid waste. These alkali-activated blends are cheap and low-carbon, but their recipes are complicated and lab data are scarce, a hard setting for prediction. The team assembled strength records from three earlier studies, used a generative adversarial network to synthesize extra realistic training data, and tuned a gradient boosting model.

The synthetic data preserved the original data's statistical patterns and correlations, steadied training, and improved generalization to unseen mixes. The tuned model predicted strength with very high accuracy on the test set, clearly leading the alternatives tried. An explanation analysis then ranked the ingredients by influence, finding cement content among the strongest levers. The pairing of synthetic data with tuned machine learning, they argue, gives engineers a reliable design tool where experiments are expensive.

Catalog record

CREATORQiang Yue, Guangming Yu, Yong Yao, Zhiqi Yang, Yufeng Li, Liyang Bai, Guangwei Xiong
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE56733
LICENSECC_BY
RIGHTSCC BY
SOURCEPLOS

Source: https://doi.org/10.1371/journal.pone.0357051

Browse the full library · About the Tower of Babel