Great!
If we could get the language detection problem sorted out, this feedback
would be highly useful.
We do have another preferred language detector, langid, with 97 supported
languages by now: https://github.com/saffsd/langid.py
best,
Gema
On Mon, Mar 9, 2015 at 9:15 AM, Kevin Brubeck Unhammer <[email protected]>
wrote:
> Per Tunedal <[email protected]>
> writes:
>
> > Hi,
> > wouldn't it be great if the input language was detected automatically?
> > Maybe TextCat http://www.let.rug.nl/vannoord/TextCat/ would do the
> > trick?
>
> It already is, using CLD2 on the server; but there are many languages
> not supported by CLD2, and it might also be trained on different types
> of text.
>
> I suppose we could try training my Python port
>
> https://github.com/unhammer/gt-CorpusTools/blob/master/corpustools/text_cat.py
> and see if it turns out to work better; we might even make it learn
> online if people choose a different source language that overrides the
> textcat suggestion :-) Or at least manually add mistakes from the top of
> the various frequency lists to the word-override lists.
>
>
> --
> Kevin Brubeck Unhammer
>
> GPG: 0x766AC60C
>
>
> ------------------------------------------------------------------------------
> Dive into the World of Parallel Programming The Go Parallel Website,
> sponsored
> by Intel and developed in partnership with Slashdot Media, is your hub for
> all
> things parallel software development, from weekly thought leadership blogs
> to
> news, videos, case studies, tutorials and more. Take a look and join the
> conversation now. http://goparallel.sourceforge.net/
> _______________________________________________
> Apertium-stuff mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/apertium-stuff
>
>
--
Gema Ramírez
---------------------
Prompsit LE
Traduce, extrae, analiza: http://aplica.prompsit.com
------------------------------------------------------------------------------
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the
conversation now. http://goparallel.sourceforge.net/
_______________________________________________
Apertium-stuff mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/apertium-stuff
