Influence of data augmentation on machine learning algorithms & predicting excavation damage zone depth
Bibliographic record
Abstract
The construction of underground projects fundamentally disturbs the natural stress equilibrium; therefore, determining the depth of damage around an underground opening is crucial. In the absence of a rich and diverse dataset, the diversity of the training data was increased by implementing data augmentation (DA) techniques, such as noise injection and scaling. Gaussian and continuous uniform distributions were used along two ranges for DA. The Gaussian distribution with the high range reduced the collinearity among the input variables (IVs) the most. After selecting the DA methods, the multiplication factors used were 2, 3, 5, and 10 times. Using a ten-fold factor reduced the variance inflation factor values the most. The machine learning algorithms' performance before and after DA was compared, and a negligible negative impact was observed. Thus, the excavation damage zone depth prediction was carried out without DA using the support vector machine (SVM). The stress component had the most influence on the SVM models. Subsequently, medium-sized and small-sized SVM models with 6 different combinations of IVs each were analyzed. It was demonstrated that the small-sized model with only three IVs, E, CI, and σ1 , showed the best performance, making it a viable option replacing the all-inclusive models.
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 | 0.005 | 0.023 |
| Meta-epidemiology (narrow) | 0.002 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.002 |
| Insufficient payload (model declined to judge) | 0.001 | 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".