Table 5

The table compares the performance of a decision tree derived using the C4.5 algorithm and a Random Forest, both generated with standard parameters. Ten-fold stratified cross-validation (on MTS) and validation on an independent protectivity validation set (MVS) are compared


C4.5
Random Forest


% sep
FPrate
TPrate
% sep
FPrate
TPrate

Cross-validation
70.41
0.24
0.65
83.39
0.15
0.82
Validation-set
73.64
0.26
0.73
84.27
0.15
0.83

Column abbreviations: %sep (% class separation), FPrate (False Positive rate) and TPrate (True Positive rate).

Sollner et al. Immunome Research 2008 4:1   doi:10.1186/1745-7580-4-1