Table 2.

Comparison performance between our model and state-of-the-art models based on Disbiome dataset.

Methods	AUC(5-fold cv)	AUC(2-fold cv)
KATZHMDA (Zhu et al., 2021) (network-based)	0.6779±0.0141	0.6696±0.0058
NTSHMDA (Luo and Long, 2018) (network-based)	0.8294±0.0071	0.8086±0.0058
NGRHMDA (Huang et al., 2017) (binary local features-based)	0.8313±0.0052	0.8233±0.0046
BiRWMP (Luo and Xiao, 2017) (binary local features-based)	0.8344±0.0089	0.8139±0.0060
GRNMFHMDA (He et al., 2018) (matrix factorization-based)	0.8609±0.0047	0.8501±0.0017
GATMDA (Long et al., 2021) (graph neural network-based)	0.9307±0.0079	0.9296±0.0154
Our model	0.9428±0.0026	0.9290±0.0068