Table 9 The comparison of R-SVM vs. WV on Data-S
From: Recursive SVM feature selection and sample classification for mass-spectrometry and microarray data
Levela | ReduceTestd | P(test-diff)e | ImproveRecf | P(rec-diff)g |
|---|---|---|---|---|
800 | -12.32% | 1.58E-04 | -4.01% | 8.77E-33 |
600 | -30.90% | 1.38E-19 | -0.16% | 0.482 |
500 | -40.09% | 2.98E-32 | -0.03% | 0.940 |
400 | -48.92% | 2.95E-37 | -2.21% | 1.29E-11 |
300 | -58.87% | 1.54E-44 | -6.81% | 2.56E-35 |
200 | -64.05% | 1.72E-48 | -13.73% | 2.35E-53 |
150 | -60.96% | 1.83E-47 | -15.52% | 2.15E-52 |
100 | -56.41% | 2.62E-49 | -19.58% | 1.29E-57 |
90 | -52.91% | 1.58E-42 | -19.14% | 1.18E-51 |
80 | -50.73% | 2.35E-41 | -19.08% | 1.31E-51 |
70 | -47.11% | 4.02E-40 | -18.27% | 6.03E-47 |
60 | -43.29% | 6.80E-38 | -17.58% | 1.18E-42 |
50 | -36.01% | 1.06E-34 | -16.35% | 1.34E-37 |