Table 20 The comparison between the PFBS-RFS-RFE and other filter ones methods
Algorithm | ACC% | NO.F | Pre | Rec | F1-score | AUC | Var. |
|---|---|---|---|---|---|---|---|
MIFS | 99.875 | 10,000 | 0.999 | 0.998 | 0.988 | 1.000 | 0.000016 |
IGF | 99.875 | 3576 | 0.999 | 0.999 | 0.998 | 1.000 | 0.000016 |
mRMR | 99.750 | 650 | 0.999 | 0.997 | 0.998 | 1.000 | 0.000028 |
CfsSubsetEval | 99.627 | 4083 | 0.998 | 0.996 | 0.997 | 1.000 | 0.000036 |
ReliefAttributeEval | 99.873 | 10,000 | 0.999 | 0.999 | 0.999 | 1.000 | 0.000031 |
OneRAttributeEval | 99.627 | 7000 | 0.998 | 0.996 | 0.997 | 0.999 | 0.000036 |
ConsistencySubsetEval | 97.380 | 3 | 0.972 | 0.970 | 0.970 | 0.994 | 0.000188 |
PCA | 99.740 | 700 | 0.999 | 0.997 | 0.998 | 0.999 | 0.000059 |
MIFS, CBF and FCBF | 99.748 | 900 | 0.999 | 0.997 | 0.998 | 1.000 | 0.000092 |
Chi-square | 99.625 | 7555 | 0.997 | 0.995 | 0.996 | 1.000 | 0.000036 |
IGF, Chi-square and Bat algorithm | 99.752 | 6483 | 0.999 | 0.997 | 0.998 | 1.000 | 0.000027 |
Proposed method (PFBS-RFS-RFE) | 100.000 | 10.000 | 1.000 | 1.000 | 1.000 | 1.000 | 0.0 |