Table 7.
Confusion matrix table of the decision tree model
| Predicted value | Actual value | |
|---|---|---|
| HSR<3.5 | HSR≥3.5 | |
| HSR<3.5 | 41 | 3 |
| HSR≥3.5 | 3 | 14 |
| Outcomes of statistics | ||
| Accuracy (95% confidence interval) | 0.90 (0.79~0.96) | |
| NIR | 0.72 | |
| P-value (Acc>NIR) | <0.001 | |
| Kappa | 0.75 | |
| McNemar’s test P-value | >0.999 | |
| Sensitivity | 0.82 | |
| Specificity | 0.93 | |
| Positive predicted value | 0.82 | |
| Negative predicted value | 0.93 | |
| Prevalence | 0.27 | |
| Detection rate | 0.22 | |
| Balanced accuracy | 0.87 | |
| Precision | 0.82 | |
| Recall | 0.82 | |
| F1 score | 0.82 | |
Confusion matrix, accuracy (95% confidence interval), kappa and McNemar’s test of the decision tree in the testing (validation) group.
HSR, health star rating; NIR, no information rate; Acc, accuracy.