AGNES-SMOTE: An Oversampling Algorithm Based on Hierarchical Clustering and Improved SMOTE

Table 5

F-measure values on each dataset with different algorithms.

Dataset	SMOTE	K-means-SMOTE	Cluster-SMOTE	AGNES-SMOTE
Ecoli	0.6237	0.6374	0.6154	0.6444
Libra	0.5926	0.5106	0.6429	0.6531
Yeast1	0.5820	0.5492	0.5828	0.5888
Haberman	0.4719	0.4361	0.4327	0.4746
Satimage	0.5812	0.5661	0.5458	0.5654
Optical_digits	0.9613	0.9661	0.9601	0.9670
Abalone	0.3945	0.0150	0.3924	0.4013
Liver	0.6075	0.5967	0.5920	0.6044
LEV	0.5103	0.4153	0.4593	0.5534