Environmental Science and Pollution Research, Journal Year: 2024, Volume and Issue: unknown
Published: July 1, 2024
Language: Английский
Environmental Science and Pollution Research, Journal Year: 2024, Volume and Issue: unknown
Published: July 1, 2024
Language: Английский
Construction and Building Materials, Journal Year: 2025, Volume and Issue: 460, P. 139835 - 139835
Published: Jan. 1, 2025
Language: Английский
Citations
5Sustainable Chemistry and Pharmacy, Journal Year: 2024, Volume and Issue: 40, P. 101611 - 101611
Published: May 27, 2024
Language: Английский
Citations
7Construction and Building Materials, Journal Year: 2024, Volume and Issue: 447, P. 138104 - 138104
Published: Sept. 4, 2024
Language: Английский
Citations
4Chemical Engineering Journal, Journal Year: 2025, Volume and Issue: unknown, P. 160920 - 160920
Published: Feb. 1, 2025
Language: Английский
Citations
0Case Studies in Construction Materials, Journal Year: 2025, Volume and Issue: unknown, P. e04492 - e04492
Published: March 1, 2025
Language: Английский
Citations
0Case Studies in Construction Materials, Journal Year: 2024, Volume and Issue: 20, P. e03157 - e03157
Published: April 18, 2024
Chloride ion is severely harmful to reinforced concrete (RC) structures in marine environments. For maintaining the durability and safety of designed RC structures, determination chloride concentration on surfaces critical. Currently, surface can be determined using empirical formulas machine learning (ML) approaches. However, these approaches only rely numerical information within field records, disregarding valuable semantic background leading low accuracy. Meanwhile, splash environments, it presents a significant challenge obtain records due complex environment high costs involved. Therefore, based limited concentrations utilizing state-of-the-art language model (LM), this study proposes an LM-based generation (LMIG) improve accuracy structures. This paper utilizes 70 sets fine-tune LMIG generates 200 high-quality records. These are then used train ML algorithms for predicting surfaces. After conducting comparative research, was found that incorporating generated by significantly enhances algorithm. Specifically, predictive random forest algorithm increased 33.1%. Furthermore, also conducts generative adversarial network (GAN)-assisted data-driven method. The results demonstrate integrating into shows advantages enhancing algorithms'
Language: Английский
Citations
3Journal of Building Engineering, Journal Year: 2024, Volume and Issue: unknown, P. 110823 - 110823
Published: Sept. 1, 2024
Language: Английский
Citations
3Corrosion Science, Journal Year: 2024, Volume and Issue: unknown, P. 112458 - 112458
Published: Sept. 1, 2024
Language: Английский
Citations
1Construction and Building Materials, Journal Year: 2024, Volume and Issue: 449, P. 138414 - 138414
Published: Sept. 26, 2024
Language: Английский
Citations
1Journal of Building Engineering, Journal Year: 2024, Volume and Issue: unknown, P. 111588 - 111588
Published: Dec. 1, 2024
Language: Английский
Citations
1