Table 3 Summary of predictive performance of each model for each emotion.

From: Personalized prediction of negative affect in individuals with serious mental illness followed using long-term multimodal mobile phenotyping

AUC

Emotion

PEM-Enet

PEM-SVM

PEM-RF

PDEM

GLMER

Anxious

0.749

0.722

0.764

0.765

0.687

Irritable

0.711

0.683

0.716

0.724

0.648

Upset

0.716

0.688

0.717

0.728

0.643

Lonely

0.774

0.739

0.787

0.791

0.690

Accuracy at Optimal Cut-offs

Anxious

0.680

0.660

0.690

0.701

0.645

Irritable

0.653

0.632

0.653

0.667

0.613

Upset

0.646

0.645

0.643

0.678

0.593

Lonely

0.709

0.678

0.710

0.719

0.644

Sensitivity at Optimal Cut-offs

Anxious

0.695

0.672

0.729

0.698

0.653

Irritable

0.661

0.633

0.665

0.658

0.596

Upset

0.669

0.605

0.666

0.639

0.621

Lonely

0.702

0.696

0.740

0.728

0.644

Specificity at Optimal Cut-offs

Anxious

0.674

0.656

0.675

0.703

0.642

Irritable

0.650

0.632

0.649

0.670

0.619

Upset

0.638

0.658

0.635

0.691

0.584

Lonely

0.712

0.671

0.699

0.715

0.644

Back to article page