Skip to main content

Table 6 Datasets, cost weights and WCAs with the two approaches proposed

From: Learning misclassification costs for imbalanced classification on gene expression data

Dataset Cost weight WCA
type p w optimal wc1 wc2 wc3 optimal wc1 wc2 wc3 ECSELM
ovarian 1.68 0.1 1.65 1.63 1.53 1.58 0.9055 0.9695 0.1966 0.2084 0.1017
Prostate 2.5 0.9 1.04 1.05 1.05 1 0.9372 0.9815 0.9509 0.9869 0.8985
Lung1 5 0.1 4.26 4.03 4.1 3.94 0.9078 0.9778 0.9786 0.9779 0.875
Lung2 8 0.9 0.92 0.9 0.66 0.61 0.9009 0.9564 0.9762 0.9675 0.9