Table 3.
The F-measures obtained by clustering different ITS datasets with the thresholds from 97 % to 100 % of the UNITE species hypotheses. The F-measure obtained by clustering the datasets with the optimal threshold of 98.41 % predicted for the CBS dataset are also given.
| Threshold | F-measure of M1 | F-measure of CT1 | F-measure of CM1 | F-measure of CC1 | F-measure of CCN1 |
|---|---|---|---|---|---|
| 97.0 % | 77.83 % | 84.44 % | 87.04 % | 86.92 % | 84.43 % |
| 97.5 % | 79.96 % | 86.34 % | 88.92 % | 88.65 % | 87.02 % |
| 98.0 % | 80.87 % | 87.75 % | 89.69 % | 89.48 % | 87.58 % |
| 98.5 % | 81.50 % | 89.74 % | 90.67 % | 90.21 % | 87.79 % |
| 99.0 % | 81.48 % | 91.47 % | 90.06 % | 89.80 % | 86.60 % |
| 99.5 % | 82.22 % | 91.40 % | 88.49 % | 88.60 % | 84.15 % |
| 98.41 % | 81.50 % | 89.74 % | 90.67 % | 90.21 % | 87.84 % |