Press Release
Free Access

Table 2.

Evaluation of the impact of the different observables in the final selection.

Precision Recall Precision Recall PMS CBe PMS CBe
PMS (%) PMS (%) CBe (%) CBe (%) (p ≥ 0.5) (p ≥ 0.5) P/R (%) P/R (%)
All observables 40.7± 1.5 78.8± 1.4 88.6± 1.1 85.5± 1.2 8470 693 100/100 100/100
No r, Hα 35.5 ± 1.4 73.1 ± 1.5 87.1 ± 1.1 85.52 ± 0.97 9108 718 77/82 80/83
No GBP, G, GRP 32.11 ± 0.99 74.2 ± 1.2 87.5 ± 1.0 82.2 ± 1.3 10772 650 63/80 80/75
No J, H, Ks 32.4 ± 1.2 81.9 ± 1.4 87.42 ± 0.84 86.59 ± 0.64 11123 758 69/91 81/88
No W1, W2, W3, W4 28.33 ± 0.68 80.4 ± 1.3 70.2 ± 1.5 76.6 ± 1.5 13303 1890 47/74 25/69
No W1, W2 37.6 ± 1.1 75.1 ± 1.3 86.0 ± 1.2 85.8 ± 1.1 8398 742 76/75 75/80
No W3, W4 31.7 ± 1.1 81.0 ± 1.2 83.9 ± 1.2 75.4 ± 1.9 11889 820 68/95 58/69
No Gvar, Vhtg 23.49 ± 0.66 74.0 ± 1.3 91.98 ± 0.83 80.7 ± 1.1 18055 358 40/84 100/52
No Gvar 27.78 ± 0.84 74.7 ± 1.4 92.62 ± 0.80 80.2 ± 1.8 13667 355 53/85 99/51
No Vhtg 40.0 ± 1.1 73.1 ± 1.4 91.89 ± 0.97 83.8 ± 1.6 7543 468 97/87 100/67

Notes. Metrics and number of sources obtained for the different classifications excluding the indicated observables. Precisions are lower limits (see Appendix B.1). Last two columns indicate the “precision” and “recall” of these new catalogues with respect to the catalogues obtained when using all the observables. This is, the proportion of sources of the different catalogues that are in the catalogues obtained using all the observables and the proportion of the candidates obtained when using all the observables that are still retrieved when excluding the respective observables.

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.