Table 3.
Average classification accuracy over 100 runs on data obtained by grouping all control and BPH samples into one class, and all cancer samples into another. Class sizes thus remain approximately balanced. Numbers in parentheses are standard deviations.
# of peaks used (malignant vs. other) | |||||
5 | 8 | 10 | 12 | 15 | |
Quadr. Disc. | 84.1 (5.3) | 85.1 (5.4) | 85.0 (6.1) | 86.1 (6.7) | 86.0 (6.1) |
Nonpar. (Kernel) | 84.6 (5.2) | 87.1 (5.3) | 88.3 (5.8) | 88.9 (6.1) | 88.1 (6.0) |
kNN | 89.9 (4.6) | 87.4 (5.6) | 87.5 (5.7) | 88.9 (5.2) | 88.5 (4.6) |
Fisher Linear | 88.6 (5.9) | 88.4 (5.6) | 87.9 (4.9) | 89.1 (5.4) | 88.0 (5.0) |
Linear SVM | 89.5 (5.5) | 91.0 (4.8) | 91.9 (4.6) | 91.7 (4.9) | 91.9 (4.7) |