From: Implementation of multiple-instance learning in drug activity prediction
Data set | Dissimilarity measure | Cross-validationa | λ | n b |
---|---|---|---|---|
I | Soergel | 0.777 | 8.000 | 196 |
 | Dice | 0.761 | 4.400 | 165 |
 | Manhattanc | 0.803 | 4.400 | 130 |
 | Rogers-Tanimoto | 0.801 | 4.000 | 153 |
II | Soergel | 0.865 | 0.001 | 103 |
 | Dice | 0.865 | 0.001 | 85 |
 | Manhattanc | 0.877 | 0.022 | 63 |
 | Rogers-Tanimoto | 0.868 | 0.069 | 72 |
III | Soergel | 0.899 | 0.001 | 94 |
 | Dice | 0.901 | 0.001 | 75 |
 | Manhattan | 0.934 | 0.550 | 63 |
 | Rogers-Tanimotoc | 0.935 | 4.400 | 46 |
IV | Soergel | 0.579 | 0.003 | 125 |
 | Dice | 0.544 | 0.031 | 111 |
 | Manhattan | 0.690 | 0.550 | 87 |
 | Rogers-Tanimotoc | 0.689 | 6.800 | 78 |