Table 8. MIA result summary of threshold attack on breast cancer dataset for heavy noise level.

Heavy noise ( $\epsilon =0.5,\delta ={10}^{-3}$) Attack type = threshold attack
Slice feature/	Attacker	AUC	Precision	Recall	F1 score
Slice value	advantage
Entire dataset	0.05	0.50	0.48	0.48	0.48
Class 0	0.09	0.48	0.49	0.48	0.49
Class 1	0.10	0.53	0.47	0.50	0.48
Correctly classified	0.06	0.49	0.49	0.48	0.48
True
Correctly classified	0.38	0.56	0.44	0.52	0.47
False