Table 7 Performance metrics of machine learning models for predicting mental and behavioral disorders.

From: Predicting mental health disparities using machine learning for African Americans in Southeastern Virginia

Disorders

 

Model Performance

Model

AUC

CA

F1

Prec

Recall

MBD

GB

0.955

0.929

0.747

0.790

0.709

LR

0.937

0.914

0.689

0.747

0.639

ANN

0.936

0.914

0.690

0.741

0.645

RF

0.925

0.919

0.699

0.774

0.637

MAD

GB

0.832

0.755

0.719

0.720

0.718

LR

0.785

0.711

0.674

0.664

0.685

ANN

0.781

0.708

0.671

0.661

0.682

RF

0.780

0.706

0.670

0.658

0.682

SSDD

GB

0.832

0.754

0.709

0.696

0.724

LR

0.785

0.713

0.659

0.649

0.669

ANN

0.781

0.709

0.651

0.647

0.656

RF

0.779

0.706

0.648

0.644

0.652

  1. GB gradient boosting, LR logistic regression, ANN artificial neural network, RF random forest, AUC area under the curve, CA correct classification, F1 F-measure or F-score (F1), (Prec) Precision, Recall sensitivity or the true positive rate (Recall).
Back to article page