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

Smart Learning Environments

Table 6 Performance of the different DM methods in predicting the academic grade using SMOTE

DM Algo	Prediction		After the 1st semester				After the 2nd semester
DM Algo	Performance measure		Precision	Recall	F1	Accuracy	Precision	Recall	F1	Accuracy
LR	Class	Above average	0.83	0.84	0.83	0.75	0.86	0.84	0.85	0.81
		Average	0.65	0.64	0.65		0.71	0.73	0.72
		Below average	0.76	0.76	0.76		0.86	0.85	0.85
RF	Class	Above average	0.83	0.84	0.84	0.82	0.89	0.85	0.87	0.87
		Average	0.75	0.72	0.74		0.80	0.81	0.81
		Below average	0.88	0.91	0.89		0.92	0.94	0.93
KNN (K = 5)	Class	Above average	0.76	0.77	0.77	0.70	0.83	0.80	0.81	0.81
		Average	0.63	0.58	0.60		0.73	0.72	0.73
		Below average	0.71	0.76	0.74		0.87	0.91	0.89
NB	Class	Above average	0.81	0.85	0.83	0.74	0.81	0.86	0.84	0.80
		Average	0.63	0.58	0.60		0.71	0.68	0.69
		Below average	0.77	0.78	0.73		0.87	0.86	0.86
SVM (poly)	Class	Above average	0.74	0.86	0.80	0.70	0.81	0.87	0.84	0.80
		Average	0.59	0.55	0.57		0.72	0.69	0.70
		Below average	0.76	0.69	0.72		0.88	0.84	0.86
ANN	Class	Above average	0.81	0.80	0.80	0.71	0.85	0.80	0.82	0.78
		Average	0.58	0.61	0.59		0.65	0.74	0.69
		Below average	0.74	0.72	0.73		0.87	0.79	0.83