Open Access

Table 5.

Scoring of the fiducial (top), optimistic, (middle) and pessimistic (bottom) models for different thresholds on p(j ∈ k|dj, {β}), as reported in the first column.

Fiducial
Thr. %TP %TN %FP %FN Precision Recall
0.1 98 83 17 2 0.19 0.98
0.2 92 89 12 8 0.25 0.92
0.5 38 98 3 62 0.43 0.38
0.7 14 100 1 87 0.90 0.14
0.9 7 100 0 93 1.00 0.07

Optimistic

Thr. %TP %TN %FP %FN Precision Recall

0.1 100 74 26 1 0.78 1.00
0.2 100 77 23 1 0.79 1.00
0.3 99 79 21 1 0.81 0.99
0.5 97 83 17 3 0.84 0.97
0.7 91 88 12 9 0.87 0.91
0.9 51 95 5 49 0.91 0.51
Pessimistic
Thr. %TP %TN %FP %FN Precision Recall
0.1 83 100 0 17 0.62 0.83
0.2 50 100 0 50 1.00 0.50
0.3 17 100 0 83 1.00 0.17
0.5 17 100 0 83 1.00 0.17
0.7 0 100 0 100 0 0
0.9 0 100 0 100 0 0

Notes. The scoring metrics are reported in Sect. 2.5.

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.