Table 12 Comparison of weighted accuracies of the LAD models constructed on three different support sets
From: Breast cancer prognosis by combinatorial analysis of gene expression data
Support set | Training set (78 cases) | Test set (19 cases) | Entire dataset (78 + 19 cases) | |
|---|---|---|---|---|
| Â | Direct classification (%) | Cross-validation (%) | Direct classification (%) | Direct classification (%) |
231 genes (van't Veer [4]) | 100.00 | 79.48 | 84.52 | 78.35 |
70 genes (van 't Veer [4]) | 99.26 | 75.43 | 84.52 | 74.06 |
Proposed support set of 17 genes | 100.00 | 82.52 | 92.86 | 81.74 |