Evolutionary Data Preprocessing to Alleviate Class Imbalance

Table 8

Performance of the regression models generated from the SVM, decision tree, and random subspace test results (10-fold cross-validation).

Classifiers	Measures	MLP regressor	SVR	Random forest
SVM	Corr.	0.974	0.334	0.967
	MAE	0.003	0.013	0.003
	RMSE	0.004	0.018	0.005
	RAE	19.67	90.04	22.59
	RRQE	22.64	95.04	26.24
Decision tree	Corr.	0.366	0.323	0.500
	MAE	0.019	0.018	0.020
	RMSE	0.028	0.029	0.026
	RAE	93.73	91.03	86.77
	RRQE	93.08	96.70	86.47
Random subspace	Corr.	0.224	0.205	0.202
	MAE	0.042	0.041	0.042
	RMSE	0.047	0.048	0.047
	RAE	96.06	93.75	95.87
	RRQE	97.46	100.46	98.78

Corr.: correlation coefficient; MAE: mean absolute error; RMSE: root mean squared error; RAE: relative absolute error; RRQE: root relative squared error.