Table 6 The classification performances of different methods for 4 gene expression datasets
From: Sparse logistic regression with a L1/2 penalty for gene selection in cancer classification
Dataset | Method | Cross-validation error | Test error | No. of selected genes |
|---|---|---|---|---|
Leukaemia | L1/2 | 2/50 | 1/22 | 2 |
LEN | 1/50 | 1/22 | 9 | |
L1 | 1/50 | 1/22 | 6 | |
Prostate | L1/2 | 5/71 | 3/31 | 5 |
LEN | 5/71 | 4/31 | 34 | |
L1 | 5/71 | 3/31 | 25 | |
Colon | L1/2 | 4/42 | 3/20 | 5 |
LEN | 5/42 | 4/20 | 13 | |
L1 | 5/42 | 4/20 | 7 | |
DLBCL | L1/2 | 3/60 | 2/17 | 14 |
LEN | 2/60 | 1/17 | 38 | |
| Â | L1 | 3/60 | 3/17 | 23 |