Table 4.

Performance of classifiers with different classification algorithms and feature extraction schemes.

Classification algorithm	Feature extraction scheme		ACC	MCC	F1-measure
SVM (polynomial kernel)	Single sampling		0.6487	0.2997	0.6252
	Multiple sampling	Discrete strategy	0.6989	0.4240	0.6357
		Continuous strategy	0.9152	0.8356	0.9101
SVM (RBF kernel)	Single sampling		0.6608	0.3276	0.6251
	Multiple sampling	Discrete strategy	0.6987	0.4188	0.6415
		Continuous strategy	0.9191	0.8428	0.9147
Adaboost M1	Single sampling		0.6693	0.3435	0.6392
	Multiple sampling	Discrete strategy	0.6574	0.3186	0.6287
		Continuous strategy	0.9024	0.8102	0.8963
Bagging	Single sampling		0.7909	0.5828	0.7848
	Multiple sampling	Discrete strategy	0.8386	0.6799	0.8317
		Continuous strategy	0.9273	0.8580	0.9240
Bayesian network	Single sampling		0.7007	0.4076	0.6722
	Multiple sampling	Discrete strategy	0.6950	0.3980	0.6614
		Continuous strategy	0.8473	0.7236	0.8225
Naive Bayes	Single sampling		0.6368	0.2822	0.5859
	Multiple sampling	Discrete strategy	0.6272	0.2616	0.5782
		Continuous strategy	0.8528	0.7329	0.8296
KNN	Single sampling		0.7652	0.5321	0.7740
	Multiple sampling	Discrete strategy	0.7918	0.5838	0.7931
		Continuous strategy	0.9071	0.8148	0.9054
Decision tree	Single sampling		0.7635	0.5315	0.7471
	Multiple sampling	Discrete strategy	0.8154	0.6333	0.8080
		Continuous strategy	0.9170	0.8359	0.9142
PART	Single sampling		0.6986	0.4015	0.6753
	Multiple sampling	Discrete strategy	0.8022	0.6105	0.7874
		Continuous strategy	0.9192	0.8402	0.9166
Logistic regression	Single sampling		0.6501	0.3008	0.6383
	Multiple sampling	Discrete strategy	0.7690	0.5442	0.7515
		Continuous strategy	0.9157	0.8353	0.9115
Multilayer perceptron	Single sampling		0.6680	0.3438	0.6352
	Multiple sampling	Discrete strategy	0.8139	0.6305	0.8052
		Continuous strategy	0.8616	0.7299	0.8688
RIPPER	Single sampling		0.7037	0.4090	0.6904
	Multiple sampling	Discrete strategy	0.7546	0.5156	0.7382
		Continuous strategy	0.9215	0.8460	0.9181