Skip to main content

Table 3 Evaluation of the classifiers KNN, SVM, RF, and NB of individuals with PD and with HY = 1

From: Discrimination between healthy and patients with Parkinson’s disease from hand resting activity using inertial measurement unit

Classifiers

Metrics

Training accuracy

sensitivity

Specificity

Precision

Accuracy

20% of features

 KNN

0.667

0.556

0.556

0.606

0.722

 SVM

0.733

0.611

0.611

0.667

0.639

 RF

0.600

0.444

0.474

0.515

0.722

 NB

0.533

0.833

0.727

0.697

0.583

40% of features

 KNN

0.733

0.556

0.579

0.636

0.778

 SVM

0.933

0.778

0.778

0.848

0.750

 RF

0.533

0.778

0.667

0.667

0.611

 NB

0.667

0.722

0.667

0.697

0.583

  1. The highest values of the metrics are highlighted in bold