Skip to main content

Table 1 Recall, precision, F-measure and rank for best gene normalization run per team

From: Overview of BioCreative II gene normalization

Team/run Recall Precision F-measure micro-average Rank micro F-measure macro-average Rank macro Significance range
T042_1 0.833 0.789 0.810 1 0.811 1 3-20
T034_1 0.815 0.792 0.804 2 0.782 2 8-20
T013_1 0.768 0.833 0.799 3 0.779 3 8-20
T004_1 0.734 0.841 0.784 4 0.777 4 8-20
T109_1 0.824 0.743 0.781 5 0.775 5 8-20
T104_1 0.743 0.807 0.774 6 0.773 6 9-20
T101_2 0.743 0.801 0.771 7 0.755 7 10-20
T107_1 0.740 0.784 0.761 8 0.739 *9 12-20
T113_2 0.761 0.752 0.756 9 0.745 *8 11-20
T108_3 0.749 0.726 0.737 10 0.724 10 13-20
T007_2 0.703 0.746 0.724 11 0.694 *12 16-20
T017_1 0.708 0.720 0.714 12 ‡0.710 *11 15-20
T110_1 0.629 0.783 0.698 13 0.685 *14 16-20
T111_3 0.664 0.717 0.689 14 0.664 *15 17-20
T030_1 0.661 0.716 0.687 15 0.649 *16 17-20
T006_2 0.606 0.767 0.677 16 ‡0.686 *13 19-20
T036_1 0.713 0.520 0.602 17 0.595 17 19-20
T014_1 0.485 0.762 0.593 18 0.584 18 20
T102_3 0.790 0.425 0.552 19 0.559 19 20
T058_2 0.415 0.375 0.394 20 0.398 20  
  1. Columns include recall, precision, F-measure and rank for micro-averaged scores, F-measure and rank for macro-averaged scores, and significance based on macro-averaged score distributions. Bold * indicates that rank for micro-average and macro-average are different; ‡ indicates that a different run was used as the high-scoring run in macro-averaged versus micro-averaged results.