Re: [idn] chinese/hangul ML.com statistics with DUDE/LDUDE

["Soobok Lee" <lsb@postel.co.kr>]

Simple RACE decoding on ML.com samples
produces the original code points, which are then fed into
DUDE.   
If you are interested in how I get the samples,
send me a personal email. (VGRS people knows that :-)).

Regards,

Soobok


----- Original Message ----- 
From: "Martin Duerst" <duerst@w3.org>
To: "Soobok Lee" <lsb@postel.co.kr>; <idn@ops.ietf.org>
Sent: Tuesday, July 10, 2001 11:43 PM
Subject: Re: [idn] chinese/hangul ML.com statistics with DUDE/LDUDE


> At 22:37 01/07/10 +0900, Soobok Lee wrote:
> >
> >The next table is from
> >285108 chinese ML.com samples (old raw data from VGRS).
> 
> Can you explain what you sampled (i.e. how you extracted
> the words,..?).
> 
> On the other hand, maybe I shouldn't ask, because this looks
> more and more like a PhD on the subject rather than like
> practical engineering.
> 
> Regards,   Martin.
> 
> 
>