[
https://issues.apache.org/jira/browse/CTAKES-196?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13658891#comment-13658891
]
Tim Miller commented on CTAKES-196:
-----------------------------------
Looks like this is not supported by OpenNLP parser API, but it would be simple
to implement. I'll try it, if it is great then I can ping the OpenNLP team and
see if they're interested in my changes.
> Create option to use POS tagger tags in constituency parser
> -----------------------------------------------------------
>
> Key: CTAKES-196
> URL: https://issues.apache.org/jira/browse/CTAKES-196
> Project: cTAKES
> Issue Type: Improvement
> Components: ctakes-constituency-parser
> Reporter: Tim Miller
> Assignee: Tim Miller
> Priority: Minor
>
> Depending on the size of training data, POS tagging may be more accurate than
> the tagging produced by the constituency parser. In this case, it may improve
> parser accuracy to pass the parser the POS tags from the cTAKES tagger as if
> they are gold standard.
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators
For more information on JIRA, see: http://www.atlassian.com/software/jira
