Table 3: Classification results obtained by RotBoost ensemble learning against typical 8 gene datasets in terms of PCA/ICA transformation methods.

DatasetICA_based RootBoostPCA_based RootBoost

Colon96.10 ± 0.5995.48 ± 0.61
CNS95.00 ± 0.2894.80 ± 0.59
Leukaemia98.77 ± 0.0398.75 ± 0.31
Breast97.88 ± 0.4594.39 ± 0.49
Lung99.54 ± 0.1198.11 ± 0.17
Ovarian99.40 ± 0.2699.82 ± 0.08o
MLL99.31 ± 0.5598.86 ± 0.23
SRBCT99.59 ± 0.1699.50 ± 0.31
Win tie loss5/2/1

Specifies that RotBoost is significantly better, and points out that RotBoost is notably worse at the significance level α = 0.05.