Table 3.
Corpus |
Mention level scores |
Document level scores |
||||
---|---|---|---|---|---|---|
Precision | Recall | F-Score | Precision | Recall | F-Score | |
Training |
0.82 (0.67) |
0.73 (0.59) |
0.77 (0.63) |
0.75 (0.44) |
0.70 (0.52) |
0.72 (0.48) |
Development |
0.90 (0.69) |
0.82 (0.64) |
0.86 (0.66) |
0.81 (0.46) |
0.74 (0.55) |
0.77 (0.51) |
Genome Biology |
0.93 (0.86) |
0.89 (0.82) |
0.91 (0.84) |
0.82 (0.62) |
0.74 (0.65) |
0.78 (0.64) |
Evaluation | 0.58 (0.49) | 0.68 (0.57) | 0.63 (0.53) | 0.65 (0.40) | 0.60 (0.44) | 0.63 (0.42) |
Lenient agreement is provided for precision, recall and F-measure, with strict agreement in brackets.