From: D-Confidence: an active learning strategy to reduce label disclosure complexity in the presence of imbalanced class distributions
Dataset
Imbalance
ff.ldc
c.ldc
dc.ldc
Best
Iris
19 %
84
61
3
dc
11 %
87
6 %
4
2 %
88
5
Vowels
21 %
99
26
23
10 %
39
35
7 %
98
69
55
3 %
102
58
74
c