HSDP: A Hybrid Sampling Method for Imbalanced Big Data Based on Data Partition

Table 4

G-means values of KNN + various sampling methods.

Dataset	Algorithm
	KNN	SMOTE + KNN	ADASYN + KNN	Borderline-SMOTE + KNN	HSDP (proposed) + KNN
Pima	0.6503	0.6673	0.6721	0.6635	0.6932
Yeast3	0.7536	0.7786	0.7798	0.7824	0.8034
Abalone19	0.3165	0.6028	0.6037	0.4366	0.5437
Segment0	0.9154	0.9379	0.9440	0.9261	0.9528
Page-blocks0	0.8153	0.8629	0.8601	0.8421	0.8757
Glass5	0.7887	0.7857	0.7898	0.7890	0.7715
Ecoli4	0.7012	0.7401	0.7928	0.7429	0.8438
Haberman	0.4502	0.4810	0.5211	0.4983	0.5201