Table 8 Holm post-hoc test results based on the MMCCs
From: Boosting methods for multi-class imbalanced data classification: an experimental review
Control method (CatBoost) | ||||
|---|---|---|---|---|
Algorithm | Z | p-value | Holm | Hypothesis (α = 0.05) |
AdaBoost.MH | 2.444 | 0.0145 | 0.0038 | Not rejected |
SAMME | 1.833 | 0.0667 | 0.0083 | Not rejected |
LogitBoost | 0.48 | 0.6311 | 0.05 | Not rejected |
GradientBoost | 1.1783 | 0.2386 | 0.0125 | Not rejected |
XGBoost | 0.9383 | 0.348 | 0.025 | Not rejected |
MEBoost | 1.9421 | 0.0521 | 0.0071 | Not rejected |
SMOTEBoost | 1.1783 | 0.2386 | 0.0166 | Not rejected |
RUSBoost | 1.9639 | 0.0495 | 0.0062 | Not rejected |
LightGBM | 2.444 | 0.0145 | 0.0041 | Not rejected |
AdaC1 | 2.1385 | 0.0324 | 0.0055 | Not rejected |
AdaC2 | 1.7893 | 0.0735 | 0.01 | Not rejected |
AdaC3 | 2.2694 | 0.0232 | 0.005 | Not rejected |
AdaCost | 2.3131 | 0.0207 | 0.0045 | Not rejected |