Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

< Back to Article

Fig 1.

Research framework for predicting the 28-day strength of cement.

More »

Fig 1 Expand

Table 1.

Statistical results of feature variables.

More »

Table 1 Expand

Table 2.

XGBoost meta-learners parameters to be optimized.

More »

Table 2 Expand

Fig 2.

Statistical results of 10-fold CV.

More »

Fig 2 Expand

Table 3.

Analysis of 10-fold CV results.

More »

Table 3 Expand

Fig 3.

Correlation between cement 28-day strength input and output.

More »

Fig 3 Expand

Table 4.

Paired t-test results of the performance differences between TF-XGBoost and XGBoost.

More »

Table 4 Expand

Fig 4.

Average multi-head attention heatmap of typical samples.

More »

Fig 4 Expand

Table 5.

Parameter settings for the meta-learner.

More »

Table 5 Expand

Table 6.

Statistics of 5-fold CV for different meta-learners.

More »

Table 6 Expand

Fig 5.

Training efficiency statistics.

More »

Fig 5 Expand

Fig 6.

Results of noise robustness verification.

More »

Fig 6 Expand

Fig 7.

Results of feature missing robustness verification.

More »

Fig 7 Expand

Fig 8.

Statistics of the 90% prediction interval results for typical types of cement.

More »

Fig 8 Expand

Table 7.

Model optimization results.

More »

Table 7 Expand

Fig 9.

Results of 25 MC-CV runs on the training and validation sets.

More »

Fig 9 Expand

Fig 10.

Performance of different models on the test set.

More »

Fig 10 Expand

Table 8.

Ablation experiment evaluation metric comparison.

More »

Table 8 Expand

Fig 11.

Shap analysis results.

More »

Fig 11 Expand

Fig 12.

SHAP dependency plot of 3-day compressive strength and other parameters.

More »

Fig 12 Expand