Table 1.
#Trees | #Folds | Noiseless | 5% Noise | 10% Noise | 20% Noise | ||||
---|---|---|---|---|---|---|---|---|---|
RF | FRF | RF | FRF | RF | FRF | RF | FRF | ||
50 | 5 | 0.037 | 0.029 | 0.039 | 0.034 | 0.045 | 0.034 | 0.063 | 0.051 |
10 | 0.036 | 0.028 | 0.034 | 0.028 | 0.043 | 0.035 | 0.060 | 0.049 | |
100 | 5 | 0.039 | 0.030 | 0.039 | 0.034 | 0.044 | 0.036 | 0.063 | 0.049 |
10 | 0.036 | 0.029 | 0.035 | 0.030 | 0.042 | 0.035 | 0.059 | 0.047 | |
150 | 5 | 0.041 | 0.034 | 0.036 | 0.030 | 0.047 | 0.037 | 0.060 | 0.049 |
10 | 0.031 | 0.027 | 0.034 | 0.029 | 0.042 | 0.034 | 0.060 | 0.047 | |
Improvement | 24% | 17% | 25% | 25% |
The different numbers of folds are used in training & test data separation. Bold values indicate the best performances.