Open Access
Table 2.
Human classifier performance.
| Percent classified as |
|||||
|---|---|---|---|---|---|
| Dataset | Size | TP | TP? | FP? | FP |
| Test data | 2000 | 53.6 | 26.5 | 11.0 | 8.9 |
| Simulated real OCs | 250 | 72.0 | 20.0 | 6.0 | 2.0 |
| Simulated fake OCs | 250 | 14.0 | 26.8 | 28.0 | 31.2 |
Notes. Results of human classification when applied to a test dataset of 2000 clusters detected by HDBSCAN in this work as well as two datasets of simulated real and fake clusters.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.