Table 7. Comparison on four multiclass tumor data sets; for each data set, 10 (8 for LeukemiaGloub) samples per class are randomly selected for training the rest are used for testing.
| Dataset name | LDA+SVM | ICA+SVM | SRC | MSRC-SVD | PFMSRC |
| SRBCT | 91.05( 4.61) |
88.72( 5.56) |
96.86( 2.64) |
97.56(
3.06)
|
96.98( 2.51) |
| ALL | 86.12( 3.81) |
91.38( 3.28) |
94.07( 2.38) |
94.07( 2.93) |
96.73(
1.68)
|
| MLLLeukemia | 93.81( 3.74) |
93.33( 5.16) |
95.36( 3.04) |
95.36( 2.84) |
95.83(
2.88)
|
| LukemiaGloub | 73.75( 5.25) |
77.50( 6.98) |
95.83(
2.14)
|
95.21( 2.35) |
94.90( 2.74) |
The average accuracy and corresponding standard deviations are reported.