This is the output of the clinical parser on a test set of 212 sentences from the clinical treebank that were NOT used in parser training nor in parser parameter optimization. Correct dependencies are marked black, incorrect dependencies are marked red. The overall labelled attachment score (proportion of tokens with the correct head word and dependency type) is 81%.

Next 25 License

1 clinicalparserout_0
2 clinicalparserout_1
3 clinicalparserout_2
4 clinicalparserout_3
5 clinicalparserout_4
6 clinicalparserout_5
7 clinicalparserout_6
8 clinicalparserout_7
9 clinicalparserout_8
10 clinicalparserout_9
11 clinicalparserout_10
12 clinicalparserout_11
13 clinicalparserout_12
14 clinicalparserout_13
15 clinicalparserout_14
16 clinicalparserout_15
17 clinicalparserout_16
18 clinicalparserout_17
19 clinicalparserout_18
20 clinicalparserout_19
21 clinicalparserout_20
22 clinicalparserout_21
23 clinicalparserout_22
24 clinicalparserout_23
25 clinicalparserout_24

Next 25

Copyright

License: The text and annotation are licensed under the Creative Commons Attribution-Share Alike 3.0 Unported License. See the home page of the corpus for details.