LDAGM: prediction lncRNA-disease asociations by graph convolutional auto-encoder and multilayer perceptron based on multi-view heterogeneous networks

Table 2 Comparison LDAGM with other classifiers, the results of LDAGM are optimal, as indicated in bold

Method	AUC	AUPR	MCC	ACC	Precision	Recall	F1-Score
SVM	0.930±0.0065	0.920±0.0136	0.862±0.0087	0.916±0.0063	0.924±0.0125	0.783±0.0036	0.856±0.0045
RF	0.864±0.0145	0.853±0.0365	0.804±0.0754	0.884±0.0452	0.836±0.0478	0.840±0.0136	0.812±0.0175
GAN	0.949±0.0025	0.952±0.0069	0.906±0.0078	0.891±0.0085	0.939±0.0069	0.848±0.0074	0.899±0.0085
XGBoost	0.934±0.0074	0.924±0.0051	0.883±0.0062	0.905±0.0084	0.925±0.0093	0.825±0.0065	0.901±0.0071
LDAGM	0.983±0.0058	0.988±0.0047	0.930±0.0233	0.941±0.0122	0.983±0.0106	0.925±0.023	0.939±0.0131

The bold number is the highest value of each column and its clarifes the superiority of our mode

ISSN: 1471-2105