Table 8:
Summary of score contributions for main effect dataset in Table 6 yielding ReliefF, SURF, SURF*, MultiSURF*, and MultiSURF scores. Illustrates why ‘far’ scoring hinders main effect detection.
| Features | RBA Scoring Schemes | |||||||
|---|---|---|---|---|---|---|---|---|
| A1 | A2 | A3 | ReliefF | SURF | SURF* | MultiSURF* | MultiSURF | |
| Farthest. Hit (same value) | 8 | 0 | 0 | − 1 | ||||
| Farthest Miss (same value) | 0 | 0 | 0 | +1 | ||||
| Nearest Hit (different value) | 0 | 0 | 0 | − 1 | − 1 | − 1 | − 1 | − 1 |
| Nearest Miss (different value) | 8 | 0 | 0 | +1 | +1 | +1 | +1 | +1 |
| Farthest Hit (different value) | 0 | 8 | 8 | +1 | ||||
| Farthest Miss (different value) | 8 | 8 | 8 | − 1 | ||||
| ReliefF Score Total | 8 | 0 | 0 | |||||
| SURF Score Total | 8 | 0 | 0 | |||||
| SURF* Score Total | 0 | 0 | 0 | |||||
| MultiSURF* Score Total | 0 | 0 | 0 | |||||
| MultiSURF Score Total | 8 | 0 | 0 | |||||