Table 10 Performance of model-based and model-free methods (using all features).

From: Model-based and Model-free Machine Learning Techniques for Diagnostic Prediction and Classification of Clinical Outcomes in Parkinson’s Disease

Method

acc

sens

spec

ppv

npv

lor

auc

Logistic Regression

0.439

0.400

0.456

0.243

0.635

−0.581

0.630

Random Forests

0.764

0.356

0.942

0.727

0.770

2.188

0.727

AdaBoost

0.703

0.333

0.864

0.517

0.748

1.156

0.695

XGBoost

0.730

0.333

0.903

0.600

0.756

1.537

0.710

SVM

0.743

0.200

0.981

0.818

0.737

2.536

0.750

Neural Network

0.655

0.444

0.748

0.435

0.755

0.863

 

Super Learner

0.723

0.289

0.913

0.591

0.746

1.445

 
Back to article page