Feature Selection and Classification of Clinical Datasets Using Bioinspired Algorithms and Super Learner
Table 3
Outline of training and testing instances of each .
Instances
WDBC dataset
SHD dataset
HCC dataset
HD dataset
VCD dataset
CHD dataset
ILP dataset
Total number of instances before SMOTE
569
270
165
155
310
303
583
Total number of instances after SMOTE
780
270
228
251
410
303
750
Number of training instances for FCSO/FKH/FBFO classifiers
468
162
137
151
246
182
450
60% of the total number of instances after SMOTE
Number of testing instances for FCSO/FKH/FBFO classifiers
312
108
91
100
164
121
300
40% of the total number of instances after SMOTE
Number of training instances for super learner
250
86
73
80
131
97
240
80% of the total testing instances for FCSO/FKH/FBFO classifiers
Number of testing instances for
62
22
18
20
33
24
60
Super learner
20% of the total testing instances for FCSO/FKH/FBFO classifiers
Each instance refers to the classification result pertaining to each instance of the testing set for FCSO, FKH, and FBFO classifiers and the class label corresponding to each instance of the testing set.