Skip to main content

Table 4 Performance evaluation of RF and other machine learning algorithm

From: Mobile APP-assisted family physician program for improving blood pressure outcome in hypertensive patients

Cross-validation

Algorithm

Performance evaluation parameters

ACC(%)

SE(%)

SPE(%)

PPV (%)

NPV (%)

AUC

K2

DT

70.12

42.10

87.52

67.66

70.86

0.6712

SVM

70.95

44.36

87.46

68.75

71.66

0.6642

NB

66.33

45.00

79.60

57.83

69.95

0.6874

RF

71.03

34.88

91.01

70.88

69.20

0.7291

K5

DT

70.14

42.09

87.50

67.57

70.89

0.6788

SVM

70.96

44.46

87.46

68.77

71.71

0.6601

NB

66.32

44.99

79.56

57.79

69.93

0.6885

RF

71.96

35.35

91.03

71.45

69.31

0.7316

K10

DT

70.11

42.11

87.56

67.76

70.90

0.6686

SVM

70.94

44.45

87.43

68.78

71.68

0.6619

NB

70.92

44.40

87.46

68.83

71.64

0.6774

RF

71.37

33.53

91.63

71.34

68.91

0.7286

  1. Abbreviation: DT Decision tree, SVM Support Vector Machine, NB Naïve Bayes, RF Random forest, ACC Accuracy, SE Sensitivity, SPE Specificity, PPV Positive predictive value, NPV Negative predictive value, AUC Area under the curve, K2 Twofold cross-validation, K5 Fivefold cross-validation, K10 Tenfold cross-validation