[
https://issues.apache.org/jira/browse/OPENNLP-1261?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=16862253#comment-16862253
]
Tim Allison commented on OPENNLP-1261:
--------------------------------------
For comparison, these are the results on the current SNAPSHOT with the 183.bin
model:
||Length||MS||Accuracy||
|10|115|0.63|
|20|121|0.85|
|30|92|0.85|
|40|115|0.93|
|50|124|0.92|
|100|279|0.96|
|150|333|0.97|
|200|433|0.98|
|500|1161|0.99|
|1000|2077|1.00|
|5000|9616|0.99|
|10000|19108|0.99|
|20000|39348|0.99|
> 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, leipzig_1000-sents.zip,
> opennlp_as_is_vs_1261.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)
