Table 17 Metrics for the performance evaluation of the classifiers when annotated using NICHD guidelines.

From: A machine learning pipeline to classify foetal heart rate deceleration with optimal feature set

Classifier

Acc.

Kappa

RMSE

ATP

AFP

A. Prec.

A. Rec.

A. F-S

Random Forest

55.67

0.218

0.544

0.557

0.324

0.569

0.557

0.563

MLP

60.82

0.300

0.429

0.608

0.289

0.606

0.608

0.606

Naïve Bayes

52.58

0.162

0.490

0.526

0.352

0.554

0.526

0.533

Simple Logistics

55.67

0.174

0.441

0.557

0.385

0.534

0.557

0.543

  1. Acc.: Accuracy, ATP: Average True Positive, AFP: Average False Positive, A. Prec.: Average Precision, A. Rec.: Average Recall, A. F-S: Average F-Score.
Back to article page