An Improved Oversampling Algorithm Based on the Samples’ Selection Strategy for Classifying Imbalanced Data

Table 1

The information of the imbalanced data sets in the experiments.

Data set ID	Data sets	#Abb	#Attr	#Min	#Maj	IR	Data source
1	Banana	Banana	2	75	2808	0.03	KEEL
2	Haberman’s Survival	Haberman	3	81	225	0.36	UCI
3	Bupa	Bupa	6	145	200	0.73	UCI
4	Appendicitis	Appendicitis	7	21	85	0.25	KEEL
5	Pima Indians Diabetes	Pima	8	268	500	0.54	KEEL
6	German Credit Data	German	20	300	700	0.43	UCI
7	Vehicle Silhouettes	Vehicle	18	199	647	0.31	KEEL
8	Led7digit	Led	7	52	448	0.12	UCI
9	Wisconsin	Wisconsin	9	241	458	0.53	UCI
10	Wine	Wine	13	48	130	0.37	UCI