Fig. 4.

Comparison of the binary classification accuracy Acc among the four algorithms, McTwo, CFS, PAM and RRF. The performance is illustrated on the two datasets a Gastric1 and b T1D, and the figures for the other datasets may be found in Additional file 1: Figure S2. The averaged value and the standard deviation of the classification Acc are calculated over the 30 runs of the 5-fold cross validations over the given dataset