Table 10 Comparison of R-SVM vs. WV on Data-R
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 | 26.23% | 6.15E-11 | -14.40% | 1.76E-72 |
600 | 21.40% | 5.51E-08 | -20.69% | 4.32E-74 |
500 | 20.28% | 1.12E-09 | -22.89% | 1.23E-75 |
400 | 18.40% | 2.70E- 10 | -25.16% | 2.38E-75 |
300 | 14.86% | 5.51E-08 | -28.52% | 1.01E-76 |
200 | 18.18% | 5.64E-07 | -26.47% | 3.99E-83 |
150 | 13.35% | 1.26E-05 | -23.87% | 4.98E-77 |
100 | 13.07% | 5.64E-07 | -18.37% | 1.21E-63 |
90 | 13.53% | 1.26E-05 | -17.91% | 1.03E-60 |
80 | 15.34% | 4.69E-06 | -16.69% | 1.16E-57 |
70 | 12.04% | 4.09E-04 | -15.49% | 4.30E-52 |
60 | 12.76% | 5.64E-07 | -13.95% | 4.61E-47 |
50 | 9.09% | 4.09E-04 | -12.68% | 3.89E-42 |
40 | 8.75% | 1.79E-03 | -10.50% | 1.00E-35 |
30 | 8.90% | 1.83E-04 | -7.77% | 2.22E-32 |