Skip to main content

Table 5 Performance of the DM methods in predicting the completion and non-completion of a degree using SMOTE

From: Predicting Master’s students’ academic performance: an empirical study in Germany

DM Algo

Prediction

After the 1st semester

After the 2nd semester

Performance measure

Precision

Recall

F1

Accuracy

Precision

Recall

F1

Accuracy

LR

Class

Completed

0.82

0.79

0.81

0.81

0.84

0.84

0.84

0.84

Not_completed

0.80

0.83

0.81

0.84

0.84

0.84

RF

Class

Completed

0.92

0.87

0.90

0.90

0.93

0.91

0.92

0.92

Not_completed

0.88

0.93

0.90

0.91

0.93

0.92

KNN (K = 5)

Class

Completed

0.77

0.76

0.77

0.77

0.85

0.83

0.84

0.84

Not_completed

0.76

0.78

0.77

0.83

0.86

0.84

NB

Class

Completed

0.82

0.78

0.80

0.81

0.84

0.84

0.84

0.84

Not_completed

0.79

0.83

0.81

0.84

0.84

0.84

SVM (poly)

Class

Completed

0.76

0.83

0.79

0.78

0.82

0.87

0.84

0.84

Not_completed

0.81

0.74

0.77

0.86

0.81

0.83

ANN

Class

Completed

0.84

0.80

0.82

0.82

0.88

0.80

0.84

0.84

Not_completed

0.81

0.85

0.83

0.81

0.89

0.85

  1. Bold values represent the results with best accuracies