Table 5 Results obtained by the various models.

From: Prediction of birthweight with early and mid-pregnancy antenatal markers utilising machine learning and explainable artificial intelligence

Classifiers

Accuracy

Precision

Recall

F1 score

Hamming loss

Jaccord score

Mathews correlation coefficient

1. RF

0.71

0.53

0.71

0.60

0.29

0.51

− 0.08

2. LR

0.58

0.65

0.58

0.60

0.41

0.44

0.12

3. DT

0.58

0.62

0.58

0.60

0.41

0.42

0.05

4. KNN

0.65

0.61

0.65

0.62

0.35

0.49

0.01

5. AdaBoost

0.77

0.73

0.77

0.72

0.22

0.61

0.33

6. CatBoost

0.73

0.66

0.73

0.65

0.27

0.54

0.11

7. LGBM

0.73

0.66

0.73

0.65

0.27

0.54

0.11

8. XG Boost

0.71

0.62

0.71

0.63

0.29

0.52

0.04

9. Stacked model

0.75

0.72

0.75

0.69

0.25

0.57

0.23

  1. Results of the performance measures of birthweight prediction of the models. RF, random forest; LR, logistic regression; DT, decision tree; LGBM, light gradient boosting machine; XG, boost extreme gradient boosting. AdaBoost showed high accuracy and precision, recall the least hamming loss, and a high Jaccord score. Mathew’s correlation is also given with each model.
Back to article page