Table 1.
Summary of data preprocessing steps.
Table 2.
Pre-processing summary of the dataset.
Table 3.
Class distribution.
Table 4.
Performance metrics.
Table 5.
Confusion matrix for XGBoost predictions.
Fig 1.
(a). Validation accuracy for XGBoost. (b). Validation Loss for XGBoost.
Fig 2.
Mean absolute SHAP values of the original categorical features.
Fig 3.
SHAP summary plot for original categorical feature.
Fig 4.
SHAP values for all one hot encoded feature.
Fig 5.
SHAP value distribution of the encoded features of “General_Plan”.