Table 3 Performance comparison under \(CV_1\).

From: Predicting lncRNA and disease associations with graph autoencoder and noise robust gradient boosting

 

Dataset

SDLDA

LDNFSGB

LDAenDL

LDA-VGHB

LDA-GARB

Precision

Dataset 1

0.8514 ± 0.0509

0.7004±0.0639

0.8764 ± 0.0493

0.8741 ± 0.0484

0.8636 ± 0.0450

Dataset 2

0.9399 ± 0.0154

0.8552 ± 0.0393

0.9391 ± 0.0290

0.9250 ± 0.0201

0.9344 ± 0.0147

Recall

Dataset 1

0.6521 ± 0.0732

0.6092 ± 0.0790

0.7019 ± 0.0639

0.7180 ± 0.0713

0.7682 ± 0.0452

Dataset 2

0.8239 ± 0.0437

0.8021 ± 0.0498

0.8304 ± 0.0523

0.8602 ± 0.0395

0.8680 ± 0.0429

Accuracy

Dataset 1

0.7799 ± 0.0341

0.6769 ± 0.0423

0.7996 ± 0.0312

0.8123 ± 0.0384

0.8282 ± 0.0338

Dataset 2

0.8857 ± 0.0283

0.8323 ± 0.0230

0.8879 ± 0.0289

0.8947 ± 0.0258

0.9036 ± 0.0281

F1-score

Dataset 1

0.7365 ± 0.0563

0.6462 ± 0.0451

0.7768 ± 0.0399

0.7852 ± 0.0412

0.8117 ± 0.0312

Dataset 2

0.8775 ± 0.0278

0.8260 ± 0.0230

0.8804 ± 0.0334

0.8908 ± 0.0227

0.8995 ± 0.0266

AUC

Dataset 1

0.8023 ± 0.0477

0.7346 ± 0.0465

0.8701 ± 0.0339

0.8814 ± 0.0425

0.9180 ± 0.0219

Dataset 2

0.9366 ± 0.0195

0.8839 ± 0.0270

0.9490 ± 0.0220

0.9541 ± 0.0200

0.9716 ± 0.0134

AUPR

Dataset 1

0.8461 ± 0.0553

0.7239 ± 0.0626

0.8903 ± 0.0273

0.8949 ± 0.0322

0.9160 ± 0.0286

Dataset 2

0.9533 ± 0.0129

0.8832 ± 0.0307

0.9582 ± 0.0167

0.9617 ± 0.0131

0.9723 ± 0.0101

  1. The best performance is denoted as bold.
Back to article page