Table 4 Performance of various machine learning algorithms using blood profile data for the classification of Breast cancer metastasis.

From: Classification and diagnostic prediction of breast cancer metastasis on clinical data using machine learning algorithms

Classification models

Before removal of outliers

After removal of outliers

Accuracy

Recall

Precision

F1 Score

AUC

Accuracy

Recall

Precision

F1 Score

AUC

Logistic regression

0.73

0.98

0.73

0.84

0.74

0.61

0.96

0.59

0.73

0.68

KNN

0.7

0.87

0.75

0.81

0.73

0.78

0.78

0.81

0.79

0.79

Decision trees

0.64

0.75

0.75

0.75

0.68

0.83

0.86

0.83

0.85

0.87

Random forest

0.73

0.98

0.74

0.84

0.74

0.83

0.89

0.81

0.85

0.85

SVM linear

0.72

1

0.72

0.84

0.74

0.55

1

0.55

0.71

0.62

Polynomial SVM

0.73

0.99

0.73

0.84

0.74

0.82

0.86

0.82

0.84

0.85

Radial SVM

0.72

1

0.72

0.84

0.74

0.55

1

0.55

0.71

0.62

Gradient boosting

0.72

0.99

0.72

0.84

0.72

0.81

0.95

0.76

0.84

0.85

XGBOOST

0.73

0.99

0.73

0.84

0.7

0.78

0.93

0.74

0.83

0.85

  1. The table represents the comparative performance of ML models before and after the removal of the outliers.
Back to article page