[jira] Commented: (NUTCH-960) Language ID - confidence factor

["Ken Krugler (JIRA)" <ji...@apache.org>]

    [ https://issues.apache.org/jira/browse/NUTCH-960?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12986827#action_12986827 ] 

Ken Krugler commented on NUTCH-960:
-----------------------------------

There are a number of Tika issues filed that relate to this. See TIKA-369, TIKA-496, TIKA-568.

> Language ID - confidence factor
> -------------------------------
>
>                 Key: NUTCH-960
>                 URL: https://issues.apache.org/jira/browse/NUTCH-960
>             Project: Nutch
>          Issue Type: Wish
>    Affects Versions: 1.2
>            Reporter: M Alexander
>
> Hi
> In JAVA implementation, what is the best way to calculate the confidence of the outcome of the language id for a given text?
> For example:
> n-gram matching / total n-gram * 100.
> when a text is passed. The outcome would be "en" with 89% confidence. What is the best way to implement this to the existig nutch language id code?
> Thanks

-- 
This message is automatically generated by JIRA.
-
You can reply to this email to add a comment to the issue online.