A Deep Learning Model for Predicting the Cement Soil Deformation Modulus
Bibliographic record
Abstract
Cement, widely used for backfill grouting in shield tunnels, plays a crucial role in maintaining the stability of tunnel structures. To enhance the prediction of cement performance, this study focuses on the elastic modulus ( E 50 ) and introduces a novel prediction model based on machine learning─the improved Convolutional Long Short-term Memory (ConvLSTM) model. The model is structured into two key components: differentiating parameter importance and extracting potential spatiotemporal order dependence among features. First, channel attention is employed to update the input of the Convolutional Long Short-term Memory model, enabling the differentiation of parameter importance. Next, the Convolutional Long Short-term Memory model extracts the potential spatiotemporal order dependence among features from the data. Finally, an attention mechanism is integrated to capture essential information. This model has undergone rigorous testing through various experiments to evaluate its predictive capabilities under different conditions. The results indicate that the maximum information coefficient algorithm effectively identifies the correlation with E 50, ranking the influencing factors as follows: strength, cement content, bentonite content, and curing time. Additionally, it was observed that while the Random Forest and Support Vector Regression models perform better with smaller data sets, the Convolutional Long Short-term Memory and Long Short-Term Memory models excel as the volume of data increases. Notably, the Convolutional Long Short-term Memory model outperforms traditional theoretical models, demonstrating higher predictive accuracy. Further experiments on different materials confirm the robust generalization ability of the Convolutional Long Short-term Memory model.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.001 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.001 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.002 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".