Figure 2

Effect of noise level. The classification error of the signal dataset (signal), the measured dataset (orgn), and the five imputed datasets. The underlying distribution parameters are SD , gene correlation , MV rate . Each panel in the figure corresponds to one combination of the feature selection methods and the classification rules, which is given by the title. The -axis labels the number of selected genes, the -axis is the noise level, and the -axis is the classification error.