Table 6.

Results on the noisy miRNA-disease datasets.

Method	AUC	AUPR
Noised DeepMDA	0.9334 ± 0.005	0.4558 ± 0.019
Noised SAE + ADA	0.8235 ± 0.012	0.1613 ± 0.015
Noised SAE + RF	0.8122 ± 0.015	0.1320 ± 0.014
Noised DRMDA	0.7757 ± 0.044	0.1290 ± 0.034
DeepMDA	0.9486* ± 0.002	0.5917* ± 0.014
SAE + ADA	0.9211 ± 0.002	0.4075 ± 0.011
SAE + RF	0.9249 ± 0.003	0.5674 ± 0.012
DRMDA	0.8812 ± 0.006	0.4614 ± 0.004

The AUC and AUPR scores are listed above. Generally, DeepMDA performed better than other three models when adding noise.