Table 9. MIA result summary of threshold attack on breast cancer dataset for moderate noise level.

Moderate noise ( $\epsilon =1,\delta ={10}^{-4}$) Attack type = threshold attack
Slice feature/	Attacker	AUC	Precision	Recall	F1 score
Slice value	advantage
Entire dataset	0.05	0.50	0.50	0.50	0.50
Class 0	0.09	0.48	0.49	0.47	0.48
Class 1	0.09	0.53	0.47	0.50	0.48
Correctly classified	0.06	0.52	0.51	0.50	0.50
True
Correctly classified	0.38	0.58	0.41	0.47	0.43
False