Heterogeneous network embedding enabling accurate disease association predictions

BMC Medical Genomics

Table 4 AUROC Score on miRNA-Disease Association Prediction

Method/Training ratio	50%	60%	70%	80%	90%
CATAPULT	0.811	0.833	0.843	0.867	0.877
HSMP	0.833	0.864	0.878	0.899	0.869
HSSVM	0.841	0.877	0.902	0.922	0.932
DeepWalk	0.498	0.511	0.534	0.611	0.677
LINE(1st+2nd)	0.780	0.795	0.829	0.813	0.804
DGI	0.501	0.483	0.496	0.516	0.512
TransE	0.473	0.477	0.481	0.469	0.464
AspEm	0.765	0.819	0.761	0.849	0.819
HeteWalk	0.937	0.951	0.953	0.946	0.969

ISSN: 1755-8794