Here is the copy of the Email I received from the student met in
Brussels after my presentation of apertium.


Date: Sun, 14 Jul 2013 22:36:29 +0200
From: liu qihao <[email protected]>
To: [email protected]
Subject: segmenteur automatique intelligence artificiel

salut


je t'envois les liens et fichiers que j'ai modifier par rapport au
logiciel de traduction automatique dont je t'ai parlé , j'avais écris
un article (avec plein de faute d'orthographes , tu vois les articles
"premier jet")) Segmenteur sémantique
http://linuxfr.org/users/eastwind/journaux/apprendre-%C3%A0-lire-et-taper-en-chinois-avec-des-outils-libres#toc_5:



et le copier / coller des liens pour télécharger les modèles sémantique :


Thanks for your interest in our C-ASSERT project.

We are glad to tell you that we have updated our alpha version of C-ASSERT to
version 1.0 recently.

==== What's new? ====

1. New Word Segementer and Parser Integration
We replaced the Chinese word segmenter and syntactic parser with GNU GPL
licensed Stanford NLP Group's Chinese Word Segementer and Syntactic Parser for
better system performance and satisfied processing speed. You will not suffer
from the pain of waiting a very long time for seeing an output any more.

2. New feature set for Chinese SRL
We equipped the BEST (to our best knowledge) features based on our research on
Chinese SRL task. Our system achieves *92.45%* F-measure when taking the
chtb100-chtb931 as training data and chtb001-chtb099 as the test data from
Chinese PropBank version 1.0.

3. UTF-8 Encoding Support
C-ASSERT uses UTF-8 as its default encoding now.

4. Scripts to generate the ".fn" formated training/test data are available.
We have provided the scripts of generating the specified ".fn" training
and test data format now. So, if you have Chinese PropBank corpus, you can use
those scripts as a start to construct your own training and test data now.

Please check C-ASSERT project home page for more details.
[ http://hlt030.cse.ust.hk/research/c-assert/ ]

Also, we provide a quick online demo for your try:
[ http://hlt030.cse.ust.hk/research/c-assert/demo.html ]


==== Where to download? ====

You can download the C-ASSERT version 1.0 related updates via the following
links:

1. The Semantic Model trained from Chinese PropBank corpus:
[ http://hlt030.cse.ust.hk/research/c-assert/ver1.0/chinese-propbank.model ]

2. The C-ASSERT Chinese ASSERT extension version 1.0:
[ http://hlt030.cse.ust.hk/research/c-assert/ver1.0/c-assert-v1.tar.gz ]


==== How to use? =====

Since we have made a lot of changes in C-ASSERT version 1.0, if you have
installed the old alpha version of C-ASSERT, you are suggested to REINSTALL
our C-ASSERT system by following the instructions listed on C-ASSERT's project
homepage.

For the installation, please refer to our project home page for a step-by-step
guideline.

Thanks,
--
C-ASSERT Maintainer


Bien à toi

Attachment: install.sh
Description: Bourne shell script

------------------------------------------------------------------------------
Introducing Performance Central, a new site from SourceForge and 
AppDynamics. Performance Central is your source for news, insights, 
analysis and resources for efficient Application Performance Management. 
Visit us today!
http://pubads.g.doubleclick.net/gampad/clk?id=48897511&iu=/4140/ostg.clktrk
_______________________________________________
Apertium-stuff mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/apertium-stuff

Reply via email to