Table 5 Performance comparison under \(CV_3.\)

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

0.8782 ± 0.0306

0.7782 ± 0.0270

0.8637 ± 0.0312

0.8597 ± 0.0269

0.8743 ± 0.0284

Dataset 2

0.9178 ± 0.0154

0.8548 ± 0.0156

0.9351 ± 0.0157

0.9270 ± 0.0143

0.9348 ± 0.0130

Recall

Dataset 1

0.7256 ± 0.0376

0.8169 ± 0.0408

0.8234 ± 0.0314

0.8388 ± 0.0332

0.8724 ± 0.0305

Dataset 2

0.8824 ± 0.0198

0.8818 ± 0.0204

0.8999 ± 0.0179

0.9088 ± 0.0169

0.9373 ± 0.0137

Accuracy

Dataset 1

0.8120 ± 0.0216

0.7916 ± 0.0256

0.8462 ± 0.0229

0.8504 ± 0.0189

0.8729 ± 0.0204

Dataset 2

0.9015 ± 0.0114

0.8658 ± 0.0127

0.9186 ± 0.0126

0.9185 ± 0.0110

0.9359 ± 0.0085

F1-score

Dataset 1

0.7939 ± 0.0260

0.7965 ± 0.0262

0.8426 ± 0.0232

0.8485 ± 0.0198

0.8728 ± 0.0204

Dataset 2

0.8996 ± 0.0119

0.8679 ± 0.0129

0.9171 ± 0.0130

0.9177 ± 0.0112

0.9359 ± 0.0085

AUC

Dataset 1

0.8774 ± 0.0200

0.8578 ± 0.0234

0.9110 ± 0.0197

0.9271 ± 0.0144

0.9459 ± 0.0109

Dataset 2

0.9560 ± 0.0081

0.9346 ± 0.0074

0.9708 ± 0.0062

0.9722 ± 0.0056

0.9790 ± 0.0051

AUPR

Dataset 1

0.8952 ± 0.0177

0.8489 ± 0.0289

0.9166 ± 0.0203

0.9364 ± 0.0157

0.9418 ± 0.0136

Dataset 2

0.9639 ± 0.0063

0.9273 ± 0.0098

0.9743 ± 0.0058

0.9761 ± 0.0051

0.9744 ± 0.0100

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