Table 2 Classification accuracy and F1-score of seven classifiers on two air compressors, before and after applying UEDSM. Results show consistent improvement in performance with UEDSM, especially for ALDNet.

From: Early detection of air leakage in IoT-connected compressors using enhanced data sampling with deep learning

Classification model

Before UEDSM

After UEDSM

Air compressor #1

Air compressor #2

Air compressor #1

Air compressor #2

Accuracy (%)

F1-score (%)

Accuracy (%)

F1-score (%)

Accuracy (%)

F1-score (%)

Accuracy (%)

F1-score (%)

ALDNet

98.55

82.60

94.68

72.06

98.69

84.00

95.54

80.10

Support vector machine

98.53

81.69

94.54

71.10

98.44

81.87

95.17

78.68

Decision tree

98.29

78.16

78.73

26.76

98.29

78.20

76.42

35.36

Random forest

98.50

82.45

93.45

62.67

98.56

82.25

94.54

73.26

Naive Bayes

97.11

72.16

93.21

74.63

95.58

64.05

90.19

68.01

K-Nearest neighbor

98.34

80.19

93.74

66.62

98.05

78.31

93.98

73.31

XGBoost

98.48

79.14

93.27

60.43

98.50

80.47

94.08

68.80

  1. Significant values are in [bold].
Back to article page