Predicting Student Academic Performance at Higher Education Using Data Mining: A Systematic Review

Table 5

Data mining techniques used in previous works.

DM task	DM technique	Studies	Number of studies (percentage of occurrence)
Classification	DT	[5, 6, 15–17, 21, 23, 25, 28, 30, 33, 35, 41, 44, 47–50, 52–56, 59]	24 (53%)
	NB	[6, 15, 17, 25, 29, 30, 32, 33, 35, 40, 42, 44, 46–49, 54–56, 60]	20 (44%)
	ANN	[6, 15, 22, 27, 29, 30, 32, 33, 35, 37, 40, 41, 44, 48, 52–56, 58]	20 (44%)
	SVM	[6, 15, 17, 25, 29, 31–33, 35, 40, 44, 52, 54, 60]	14 (31%)
	KNN	[29–33, 40, 42, 49, 54, 55, 60]	11 (24%)
	RF	[6, 22, 26, 29–32, 34, 42, 48, 52, 53, 55, 56]	14 (31%)
	LR	[6, 29, 32, 33, 42, 48, 54, 56, 60]	9 (20%)
	Ensemble	[17, 22, 25, 30, 32, 41, 48, 54, 56]	9 (20%)
	Bayesian network (BN)	[28]	1 (2%)
	eXtreme gradient boosting	[22, 27]	2 (4%)
	AdaBoost	[31]	1 (2%)
	Gradient boosted trees	[22, 27, 31]	3 (7%)
	ExtraTree	[31]	1 (2%)
	SMO	[33, 55]	2 (4%)
	Linear discriminant analysis	[40]	1 (2%)
	NNge	[41]	1 (2%)
Regression	SMOReg (SVM)	[36, 43]	2 (4%)
	Simple logistic regression (SLR)	[36]	1 (2%)
	DT	[36]	1 (2%)
	RF	[36, 43]	2 (4%)
	Linear regression	[36, 43, 51]	3 (7%)
	KNN	[36, 43]	2 (4%)
	ANN	[36]	1 (2%)
	Gaussian	[36, 43]	2 (4%)
	Processes random tree
	Ensemble (bagging)	[43]	1 (2%)
	M5	[43]	1 (2%)
	M5 rules	[43]	1 (2%)
	Collaborative filtering (CF)	[45]	1 (2%)
	Matrix factorization (MF)	[45]	1 (2%)
	Singular value decomposition (SVD)	[45]	1 (2%)
	Restricted Boltzmann machines (RBM)	[45]	1 (2%)
Clustering	K-means	[21, 35, 54]	3 (7%)