. 2024 Jun 20;12(6):1371. doi: 10.3390/biomedicines12061371

Table 6.

Levels of significance (p-values) obtained from the statistical analysis of the difference between the mean AUCs found for each type of classification (binary and multi-classification). p < 0.05 is represented in bold: there is a significant difference between classifications.

	Imbalanced Data		Oversampling		Undersampling
	without DA	with DA	without DA	with DA	without DA	with DA
Luminal A vs. Non-Luminal A	0.843	0.442	0.562	0.001	0.963	0.225
Luminal A (multi-classification)	0.843	0.442	0.562	0.001	0.963	0.225
Luminal B1 vs. Non-Luminal B1	0.532	0.113	0.741	0.009	0.923	0.899
Luminal B1 (multi-classification)	0.532	0.113	0.741	0.009	0.923	0.899
Luminal B2 vs. Non-Luminal B2	0.470	0.971	0.407	0.691	0.184	0.007
Luminal B2 (multi-classification)	0.470	0.971	0.407	0.691	0.184	0.007
HER2 vs. Non-HER2	0.049	0.112	0.082	0.655	0.326	0.138
HER2 (multi-classification)	0.049	0.112	0.082	0.655	0.326	0.138
TN vs. Non-TN	0.600	0.798	0.488	0.929	0.661	0.311
TN (multi-classification)