Skip to main content

Advances, Systems and Applications

Table 6 This table presents the performance results of each predictive model for the task level based on the training dataset, where the model in bold represents the best model

From: Cloud failure prediction based on traditional machine learning and deep learning

Model

Accuracy(%)

Error rate(%)

Precision(%)

Sensitivity(%)

Specificity(%)

F-Score

Logistic Regression

69.68

30.32

71.09

79.95

55.67

0.7526

Decision Tree

89.75

10.25

85.44

98.56

78.42

0.9153

Random Forest

92.47

7.53

90.66

99.13

78.86

0.9471

Gradient Boosting

87.81

12.19

84.81

95.92

76.86

0.9003

XGBoost

89.34

10.66

85.04

98.30

77.89

0.9119

Single Layer LSTM

87.74

12.26

83.94

96.85

75.87

0.8994

Bi-Layer LSTM

86.93

13.07

81.36

98.18

73.84

0.8898

Ti-Layer LSTM

87.54

12.46

82.90

79.53

75.27

0.8962