What does this research mean for the field?

An integrated machine learning framework combining ASMOTE, PINO, and SABO accurately predicts soil cohesion and internal friction angle while ensuring physical consistency and spatial transferability. Novelty: ClaimNovelty.METHODOLOGICAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

Develop an integrated ASMOTE–PINO–SABO machine learning framework to reliably predict soil cohesion and internal friction angle while resolving data imbalance and spatial transferability issues.

September 14, 2026International Journal of Construction Management

Data-driven geotechnical characterization and prediction of mechanical properties using machine learning framework

View Full Paper

Ask AI

Bookmark

Authors

SSSenthamil Kumar S

Discussion

Member takes

Overview

Computational study demonstrates accurate prediction of soil cohesion and friction angles across borehole datasets, indicating robust spatial transferability for geotechnical engineering.

Key Points

Develop an integrated ASMOTE–PINO–SABO machine learning framework to reliably predict soil cohesion and internal friction angle while resolving data imbalance and spatial transferability issues.
Integrated Adaptive Synthetic Minority Oversampling Technique (ASMOTE), Physics-Informed Neural Operator (PINO), and Self-Adaptive Bayesian Optimization (SABO) using a balanced dataset of 3,250 samples.
Implemented a leakage-free GroupKFold spatial validation across unseen borehole locations up to 87.1 km away and conducted finite element verification in ABAQUS.
Achieved R² values of 0.894 (RMSE = 4.12 kPa) for cohesion and 0.918 (RMSE = 1.21°) for friction angle, with the PINO component reducing physics residual errors by 91.05%.
Demonstrated transferability at an 87.1 km spatial distance with R² values of 0.854 and 0.882, while finite element settlement prediction errors remained below 2.1%.

Cite This Study

Senthamil Kumar S (2026) studied this question.

synapsesocial.com/papers/6aa7b3770926e14a848b278b https://doi.org/10.1080/15623599.2026.2721504

View Full Paper

Ask AI

Bookmark

September 14, 2026International Journal of Construction Management

Data-driven geotechnical characterization and prediction of mechanical properties using machine learning framework

View Full Paper

Ask AI

Bookmark

Authors

SSSenthamil Kumar S

Discussion

Member takes

Overview

Computational study demonstrates accurate prediction of soil cohesion and friction angles across borehole datasets, indicating robust spatial transferability for geotechnical engineering.

Key Points

Develop an integrated ASMOTE–PINO–SABO machine learning framework to reliably predict soil cohesion and internal friction angle while resolving data imbalance and spatial transferability issues.
Integrated Adaptive Synthetic Minority Oversampling Technique (ASMOTE), Physics-Informed Neural Operator (PINO), and Self-Adaptive Bayesian Optimization (SABO) using a balanced dataset of 3,250 samples.
Implemented a leakage-free GroupKFold spatial validation across unseen borehole locations up to 87.1 km away and conducted finite element verification in ABAQUS.
Achieved R² values of 0.894 (RMSE = 4.12 kPa) for cohesion and 0.918 (RMSE = 1.21°) for friction angle, with the PINO component reducing physics residual errors by 91.05%.
Demonstrated transferability at an 87.1 km spatial distance with R² values of 0.854 and 0.882, while finite element settlement prediction errors remained below 2.1%.

Cite This Study

Senthamil Kumar S (2026) studied this question.

synapsesocial.com/papers/6aa7b3770926e14a848b278b https://doi.org/10.1080/15623599.2026.2721504

Also Consider

Synapse has enriched 3 closely related papers on similar clinical questions. Consider them for comparative context:

View Full Paper

Ask AI

Bookmark