Table 7 Post-hoc test of stacking vs. voting for both datasets.

From: Ensemble learning with explainable AI for improved heart disease prediction based on multiple datasets

Metric

Test results

D1

D2

Accuracy

Statistic

1.5492

1.5492

Adjusted p-value

0.1213

0.1213

H0 accepted

Yes

Yes

Precision

Statistic

1.5492

1.5492

Adjusted p-value

0.1213

0.1213

H0 accepted

Yes

Yes

Recall

Statistic

1.5492

1.5492

Adjusted p-value

0.1213

0.1213

H0 accepted

Yes

Yes

F1-score

Statistic

1.5492

1.5492

Adjusted p-value

0.1213

0.1213

H0 accepted

Yes

Yes

Kappa

Statistic

1.5492

1.5492

Adjusted p-value

0.1213

0.1213

H0 accepted

Yes

Yes

MCC

Statistic

1.5492

1.5492

Adjusted p-value

0.1213

0.1213

H0 accepted

Yes

Yes

Back to article page