Skip to main content

Table 2 Binary classification metrics of different methods on Dataset2

From: A representation learning model based on variational inference and graph autoencoder for predicting lncRNA-disease associations

Sp Method Sn Acc Pre F1 Mcc
0.95 LRLSLDA 0.4572 0.9369 0.2051 0.2831 0.2777
SIMCLDA 0.2128 0.9299 0.1066 0.1421 0.1169
TPGLDA 0.5565 0.9394 0.2384 0.3338 0.3380
SKFLDA 0.5284 0.9385 0.2286 0.3191 0.3206
GAMCLDA 0.6377 0.9415 0.2635 0.3729 0.3855
VGAELDA 0.9329 0.9495 0.3434 0.5020 0.5490
0.99 LRLSLDA 0.1591 0.9676 0.3145 0.2113 0.2086
SIMCLDA 0.1020 0.9658 0.2223 0.1398 0.1348
TPGLDA 0.2673 0.9703 0.4279 0.3291 0.3238
SKFLDA 0.2354 0.9694 0.3976 0.2958 0.2913
GAMCLDA 0.4472 0.9752 0.5558 0.4956 0.4860
VGAELDA 0.7831 0.9843 0.6868 0.7318 0.7254
  1. The bold number is the highest value of each column, which is achieved by our method, VGAELDA. The bold clarifies the superiority of our method
  2. Sp specificity, Sn sensitivity, Acc accuracy, Pre precision, F1 F1-score, Mcc Matthews correlation coefficient
\