Predicting potential miRNA-disease associations based on more reliable negative sample selection

BMC Bioinformatics

Table 3 Performance comparison of existing prediction methods based on fivefold cross-validations

method	AUC	AUPR	Precision	Recall	F1-score	Accuracy
KR-NSSM
RFMDA	0.9414	0.9606	0.9818	0.8424	0.9064	0.9134
IRFMDA	0.9671	0.9739	0.9591	0.8582	0.9054	0.9109
ABMDA	0.9732	0.9789	0.9971	0.8427	0.9129	0.9201
GBDT-LR	0.9633	0.9730	0.9625	0.8654	0.9111	0.9158
SMALF	0.9913	0.9931	0.9749	0.9507	0.9626	0.9648
Original selection
RFMDA	0.7388	0.7034	0.6253	0.9548	0.7453	0.6912
IRFMDA	0.9267	0.9222	0.8447	0.8598	0.8521	0.8567
ABMDA	0.8841	0.8807	0.8152	0.7827	0.7908	0.8027
GBDT-LR	0.9274	0.9014	0.8315	0.8273	0.8302	0.8304
SMALF	0.9503	0.9472	0.8808	0.8931	0.8868	0.8860

ISSN: 1471-2105