Skip to main content

Table 5 Prediction performance using three-way data split

From: Predicting diabetic retinopathy and identifying interpretable biomedical features using machine learning algorithms

 

Training

Validation

Test

Model

Acc.

AUC

Acc.

AUC

Acc.

AUC

SVM

0.863

0.961

0.822

0.801

0.817

0.744

LR

0.831

0.769

0.813

0.707

0.798

0.712

ANN

0.872

0.849

0.794

0.707

0.780

0.685

DT

0.825

0.707

0.817

0.693

0.780

0.640

  1. Accuracy and AUC of different machine learning algorithms using training (i.e., 60%), validation (i.e., 20%), and test (i.e., 20%) data sets are evaluated
  2. aBest evaluation measures in validation set are underlined as selected mode and independent performance evaluation is shown in bold
Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us