Fig. 4From: Data reduction techniques for highly imbalanced medicare Big DataMethodology for scenario two experiments, feature selection onlyBack to article page