Skip to main content

Table 4 G-mean for our proposed method (LVQ-SMOTE) in case MWMOTE instead of SMOTE is used in our algorithm

From: LVQ-SMOTE – Learning Vector Quantization based Synthetic Minority Over–sampling Technique for biomedical data

 

Nothing: baseline

Our proposed method

Datasets

SMOTE

MWSMOTE

SMOTE

MWSMOTE

Breast-w

70.31%

70.59%

71.03%

70.69%

Blood

96.41%

96.50%

97.02%

96.40%

Colon-cancer

71.82%

71.08%

78.86%

79.09%

Ionosphere

85.86%

85.92%

89.48%

91.28%

Leukemia

95.79%

95.92%

100.0%

100.0%

Pima

75.18%

74.07%

75.73%

75.69%

Satimage

73.64%

73.92%

75.71%

77.01%

Yeast

80.66%

81.20%

81.27%

81.38%

  1. The algorithm used in this experiment is Logistic Tree.