Table 12 Performance of LDA-GARB with \(k=64\) and different N on two datasets under \(CV_3\).

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

 

Dataset

\(k=64,\,N=1\)

\(k=64,\,N=2\)

\(k=64,\,N=3\)

\(k=64,\,N=4\)

\(k=64,\,N=5\)

Precision

Dataset 1

0.8743 ± 0.0284

0.8714 ± 0.0289

0.8704 ± 0.0269

0.8711 ± 0.0274

0.8835 ± 0.0260

Dataset 2

0.9348 ± 0.0130

0.9413 ± 0.0132

0.9336±0.0138

0.9327 ± 0.0125

0.9343 ± 0.0104

Recall

Dataset 1

0.8724 ± 0.0305

0.8689 ± 0.0309

0.8781 ± 0.0346

0.8767 ± 0.0292

0.8806 ± 0.0343

Dataset 2

0.9373 ± 0.0137

0.9407 ± 0.0132

0.9335 ± 0.0134

0.9329 ± 0.0160

0.9365 ± 0.0147

Accuracy

Dataset 1

0.8729 ± 0.0204

0.8698 ± 0.0208

0.8733 ± 0.0225

0.8731 ± 0.0223

0.8819 ± 0.0225

Dataset 2

0.9359 ± 0.0085

0.9409 ± 0.0090

0.9335 ± 0.0096

0.9327 ± 0.0094

0.9353 ± 0.0090

F1-score

Dataset 1

0.8728 ± 0.0204

0.8696 ± 0.0208

0.8737 ± 0.0231

0.8735 ± 0.0221

0.8816 ± 0.0231

Dataset 2

0.9359 ± 0.0085

0.9409 ± 0.0090

0.9335 ± 0.0096

0.9327 ± 0.0095

0.9353 ± 0.0091

AUC

Dataset 1

0.9459 ± 0.0109

0.9462 ± 0.0133

0.9479 ± 0.0134

0.9367 ± 0.0174

0.9409 ± 0.0174

Dataset 2

0.9790 ± 0.0051

0.9824 ± 0.0034

0.9801 ± 0.0046

0.9790 ± 0.0048

0.9798 ± 0.0050

AUPR

Dataset 1

0.9418 ± 0.0136

0.9449 ± 0.0161

0.9432 ± 0.0192

0.9331 ± 0.0206

0.9371 ± 0.0214

Dataset 2

0.9744 ± 0.0100

0.9818 ± 0.0040

0.9787 ± 0.0061

0.9781 ± 0.0060

0.9790 ± 0.0065

  1. The best performance is denoted as bold.
  2. The second-best performance is denoted as underline.
Back to article page