[
https://issues.apache.org/jira/browse/OPENNLP-333?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=13150893#comment-13150893
]
Joern Kottmann commented on OPENNLP-333:
----------------------------------------
This depends on the training data, and is not considered to be a bug, or issues
with the OpenNLP name finder. I suggest to update and extend your training
data, or in case you are using the models from our website produce your own
training data (or buy some training data).
Please come to the user mailing list if you have further question or in case
you need some advice on how (and which data) to train the name finder.
> NameFinderME not working for single name at beginning of sentence
> -----------------------------------------------------------------
>
> Key: OPENNLP-333
> URL: https://issues.apache.org/jira/browse/OPENNLP-333
> Project: OpenNLP
> Issue Type: Bug
> Components: Name Finder
> Affects Versions: tools-1.5.1-incubating
> Reporter: Josh Smith
>
> For sentences that begin with a single name (ie first/last name only),
> NameFinderME returns an empty array of Spans. If the second word is also part
> of the name, the Span is returned properly. If the sentence is reversed such
> that the name is last, the Span is returned correctly.
> For example,
> "John is my name" -> []
> "John Doe is my name" -> [0..2]
> "My name is John" -> [3..4]
--
This message is automatically generated by JIRA.
If you think it was sent incorrectly, please contact your JIRA administrators:
https://issues.apache.org/jira/secure/ContactAdministrators!default.jspa
For more information on JIRA, see: http://www.atlassian.com/software/jira
