[
https://issues.apache.org/jira/browse/TIKA-3147?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=17165942#comment-17165942
]
Hudson commented on TIKA-3147:
------------------------------
SUCCESS: Integrated in Jenkins build tika-main-jdk8 #1835 (See
[https://builds.apache.org/job/tika-main-jdk8/1835/])
TIKA-3147 -- drop tokens below quant value. (tallison:
[https://github.com/apache/tika/commit/ca4852db326445059cdd444dcda0d5d2ac414e23])
* (edit)
tika-eval/src/main/java/org/apache/tika/eval/textstats/TextProfileSignature.java
* (edit)
tika-eval/src/test/java/org/apache/tika/eval/textstats/TextStatsTest.java
> Strip punctuation in lang id component within tika-eval
> -------------------------------------------------------
>
> Key: TIKA-3147
> URL: https://issues.apache.org/jira/browse/TIKA-3147
> Project: Tika
> Issue Type: Task
> Reporter: Tim Allison
> Priority: Major
> Fix For: 1.25
>
>
> I noticed that "the quick brown fox jumped over the lazy dog" was identified
> as English in tika-eval. However, if I added semi-colons, it was identified
> as Chinese.
> This is in alignment with what I've recently seen on a new batch of technical
> documents that are mostly numbers and abbreviations...these are being
> identified as Chinese.
> Ideally, we'd strip out punctuation while building the models, and while
> running language id.
> As a short term fix, we should at least strip out punctuation during
> detection.
--
This message was sent by Atlassian Jira
(v8.3.4#803005)
