Table 5 Optimal weights for different data set
From: Learning misclassification costs for imbalanced classification on gene expression data
Data set | Sample categorical | Influence factors | Optimal weights wc | WCA | |
|---|---|---|---|---|---|
distribution p | w1/(w1+w2) | w2/(w1+w2) | |||
Colon | 1 | 0.2 | 0.8 | 1.03 | 0.6167 |
Leukemia | 1.33 | 0.9 | 0.1 | 0.9 | 0.9179 |
Ovarian1 | 1.68 | 0.9 | 0.1 | 1.65 | 0.9055 |
Prostate1 | 2 | 0.9 | 0.1 | 1.06 | 0.939 |
Prostate2 | 2.5 | 0.9 | 0.1 | 1.04 | 0.9372 |
Lung1 | 3 | 0.9 | 0.1 | 0.93 | 0.92 |
Ovarian2 | 4 | 0.1 | 0.9 | 3.45 | 0.9094 |
Lung2 | 5 | 0.1 | 0.9 | 4.26 | 0.9078 |
Ovarian3 | 6.5 | 0.9 | 0.1 | 0.8 | 0.9075 |
Lung3 | 8 | 0.9 | 0.1 | 0.92 | 0.9009 |