[jira] [Comment Edited] (OPENNLP-1261) Language Detector fails to predict language on long input texts

["Tim Allison (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/OPENNLP-1261?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16851085#comment-16851085 ] 

Tim Allison edited comment on OPENNLP-1261 at 5/29/19 5:34 PM:
---------------------------------------------------------------

None of the above is a showstopper from my perspective.  And do beware that the lengths differ across the 2 runs.

Please let me know if you'd like me to re-run any experiments and/or if I can improve the output/calculations/methodology.

Thank you!



was (Author: tallison@mitre.org):
None of the above is a showstopper from my perspective.  Please let me know if you'd like me to re-run any experiments and/or if I can improve the output/calculations/methodology.

Thank you!


> Language Detector fails to predict language on long input texts
> ---------------------------------------------------------------
>
>                 Key: OPENNLP-1261
>                 URL: https://issues.apache.org/jira/browse/OPENNLP-1261
>             Project: OpenNLP
>          Issue Type: Improvement
>          Components: Language Detector
>            Reporter: Joern Kottmann
>            Assignee: Joern Kottmann
>            Priority: Major
>         Attachments: langid_plus_minus_rollups.zip
>
>
> If the input text is very long, e.g. 100k chars, then the lang detect component fails to detect the language correctly, even though the text is only written in one language.
> This issue was tracked down to the context generator, where the count of the ngrams are ignored.



--
This message was sent by Atlassian JIRA
(v7.6.3#76005)