Table 11 Performance of LDA-GARB with \(k=64\) and different N on two datasets under \(CV_2\).
From: Predicting lncRNA and disease associations with graph autoencoder and noise robust gradient boosting
| Â | Dataset | \(k=64,\,N=1\) | \(k=64,\,N=2\) | \(k=64,\,N=3\) | \(k=64,\,N=4\) | \(k=64,\,N=5\) |
|---|---|---|---|---|---|---|
Precision | Dataset 1 | 0.8724 ± 0.0365 | 0.8686 ± 0.0332 | 0.8897 ± 0.0272 | 0.8747 ± 0.0248 | 0.8980 ± 0.0278 |
Dataset 2 | 0.9321 ± 0.0277 | 0.9339 ± 0.0228 | 0.9264 ± 0.0269 | 0.9381 ± 0.0225 | 0.9379 ± 0.0268 | |
Recall | Dataset 1 | 0.8699 ± 0.0377 | 0.8784 ± 0.0393 | 0.8764 ± 0.0317 | 0.8535 ± 0.0455 | 0.8630 ± 0.0422 |
Dataset 2 | 0.9409 ± 0.0262 | 0.9448 ± 0.0233 | 0.9411 ± 0.0237 | 0.9357 ± 0.0235 | 0.9343 ± 0.0251 | |
Accuracy | Dataset 1 | 0.8744 ± 0.0255 | 0.8760 ± 0.0210 | 0.8855 ± 0.0221 | 0.8689 ± 0.0243 | 0.8845 ± 0.0239 |
Dataset 2 | 0.9409 ± 0.0158 | 0.9432 ± 0.0136 | 0.9371 ± 0.0149 | 0.9409 ± 0.0113 | 0.9403 ± 0.0131 | |
F1-score | Dataset 1 | 0.8707 ± 0.0316 | 0.8729 ± 0.0292 | 0.8826 ± 0.0227 | 0.8634 ± 0.0298 | 0.8795 ± 0.0282 |
Dataset 2 | 0.9363 ± 0.0243 | 0.9392 ± 0.0209 | 0.9336 ± 0.0225 | 0.9368 ± 0.0203 | 0.9359 ± 0.0219 | |
AUC | Dataset 1 | 0.9493 ± 0.0160 | 0.9516 ± 0.0137 | 0.9558 ± 0.0138 | 0.9340 ± 0.0202 | 0.9474 ± 0.0164 |
Dataset 2 | 0.9817 ± 0.0083 | 0.9824 ± 0.0063 | 0.9800 ± 0.0066 | 0.9817 ± 0.0066 | 0.9828 ± 0.0064 | |
AUPR | Dataset 1 | 0.9415 ± 0.0228 | 0.9468 ± 0.0219 | 0.9548 ± 0.0160 | 0.9316 ± 0.0219 | 0.9475 ± 0.0194 |
Dataset 2 | 0.9757 ± 0.0176 | 0.9774 ± 0.0146 | 0.9752 ± 0.0126 | 0.9754 ± 0.0202 | 0.9793 ± 0.0143 |
