Moses-support Digest, Vol 94, Issue 23

Send Moses-support mailing list submissions to
moses-support@mit.edu

To subscribe or unsubscribe via the World Wide Web, visit
http://mailman.mit.edu/mailman/listinfo/moses-support
or, via email, send a message with subject or body 'help' to
moses-support-request@mit.edu

You can reach the person managing the list at
moses-support-owner@mit.edu

When replying, please edit your Subject line so it is more specific
than "Re: Contents of Moses-support digest..."


Today's Topics:

1. CfP: AMTA WS on Interactive and Adaptive MT (Philipp Koehn)
2. translation quality (Sehrob Ibrohimov)
3. Re: translation quality (Read, James C)


----------------------------------------------------------------------

Message: 1
Date: Mon, 18 Aug 2014 14:02:14 -0400
From: Philipp Koehn <pkoehn@inf.ed.ac.uk>
Subject: [Moses-support] CfP: AMTA WS on Interactive and Adaptive MT
To: "corpora@uib.no" <CORPORA@uib.no>, mt-list@eamt.org,
"moses-support@mit.edu" <moses-support@mit.edu>, Multiple recipients
of list <mt_list@nist.gov>
Message-ID:
<CAAFADDCrOwUN_0bFsXt9poXAdyFMRgBV1Rsz_ZfnAR2TORP5bA@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"

Final Call for Papers:

AMTA 2014 WORKSHOP ON INTERACTIVE AND ADAPTIVE MACHINE TRANSLATION
22 October 2014, Vancouver, Canada

http://www.statmt.org/iamt/
Paper submission deadline: August 29, 2014

The increasing use of machine translation in the workflow of
professional translators creates demand for machine translation
technology that provides more interactive collaboration, learns from
its errors and adapts to the translators' style and adapts the
underlying machine translation system online to the specific needs of
the translator for the given task.

The next generation of computer aided translation (CAT) tools has to
move beyond the use of static machine translation for human
post-editing into a much richer division of labor between man and
machine that takes full advantage of man's understanding of content
and machine's greater ability to quickly process large amounts of
data.

On the other hand, these tools will allow a more friendly interaction
between the human and the machine through the use of different
modalities of interactions as speech, gaze tracking, e-pen, etc.
Finally, all of these issues will lead to an increase of the
productivity of the professional translators.

Such tools are in development in a number of research labs across the
world, one example is the open source workbench developed by the
EU-funded projects Matecat and Casmacat, led by the organizers of this
workshop.

This workshop brings to together researchers in this nascent subfield
of machine translation. The workshop will divide its schedule about
equally between invited talks by leading researchers and paper
presentations on more recent advances.

We encourage submissions including, but not limited to:

- rapid user and project adaptation
- online learning methods
- active learning
- domain adaptation
- interactive translation prediction
- use of confidence measures to inform translators
- quality estimation to filter/rank translation suggestions
- automatic terminology support
- integration of translation memory and machine translation
- computer assisted writing in the context of translation
- novel types of assistance
- multimodal interaction: speech, writing, gestures, gaze.
- user studies on advanced computer aided translation

INVITED TALKS

- Michael Denkowski, CMU
- Marcello Federico, Fondazione Bruno Kessler
- Jes?s Gonz?lez-Rubio, Universitat Polit?cnica de Val?ncia
- Spence Green, Stanford
- John Moran, CNGL
- Lane Schwartz, Air Force Research Laboratory
- Michel Simard, National Research Council Canada

IMPORTANT DATES

Paper submission deadline: August 29, 2014
Notification of acceptance: September 12, 2014
Camera-ready deadline: September 26, 2014

ORGANIZERS

Francisco Casacuberta, Universitat Polit?cnica de Val?ncia
Marcello Federico, Fondazione Bruno Kessler
Philipp Koehn, University of Edinburgh / Johns Hopkins University
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://mailman.mit.edu/mailman/private/moses-support/attachments/20140818/d72cb6a6/attachment-0001.htm

------------------------------

Message: 2
Date: Tue, 19 Aug 2014 09:47:13 +0500
From: Sehrob Ibrohimov <isehrob@gmail.com>
Subject: [Moses-support] translation quality
To: moses-support <moses-support@mit.edu>
Message-ID:
<CADJquiCzOxx9_sJ6cnheuucpdHVcLk+ESVbSeSZPpgAQQ=hOaw@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8

Hi everybody !

I have trained a translation model with moses (with mgiza) but not
satisfied with the quality of translation. there are about 137000
sentence pairs in my corpora, mainly legal acts. I had trained a
translation model with moses before (with GIZA++) with less sentence
pairs, about 50000, but translation was better than this. I guess that
I have some problems with my corpora, but can't find out what exactly
it is.
And also while training I got this message, "nohup: ignoring input and
redirecting stderr to stdout"
this was my command:
"nohup nice ~/mosesdecoder/scripts/training/train-model.perl -root-dir
train -corpus ~/corpus/my_corpus -f fr -e en -alignment
grow-diag-final-and -reordering msd-bidirectional-fe -lm
0:6:$HOME/lm/my_language_model:8 -mgiza -external-bin-dir
~/mosesdecoder/tools > training.out "
Could you, please, tell me if you have any idea about this problem?

thanks in advance !

Sehrob


------------------------------

Message: 3
Date: Tue, 19 Aug 2014 07:03:15 +0000
From: "Read, James C" <jcread@essex.ac.uk>
Subject: Re: [Moses-support] translation quality
To: Sehrob Ibrohimov <isehrob@gmail.com>, moses-support
<moses-support@mit.edu>
Message-ID:
<F00840E41983C645928E21E3C35F4EB101881E24AD@mbx1-node2.essex.ac.uk>
Content-Type: text/plain; charset="us-ascii"

add 2> output-file to your command and your STDERR messages will be dumped to this file instead of output to STDOUT.

I have no idea why you are not getting the results you are expecting. I expect you will have to give more information.

James

________________________________________
From: moses-support-bounces@mit.edu [moses-support-bounces@mit.edu] on behalf of Sehrob Ibrohimov [isehrob@gmail.com]
Sent: 19 August 2014 05:47
To: moses-support
Subject: [Moses-support] translation quality

Hi everybody !

I have trained a translation model with moses (with mgiza) but not
satisfied with the quality of translation. there are about 137000
sentence pairs in my corpora, mainly legal acts. I had trained a
translation model with moses before (with GIZA++) with less sentence
pairs, about 50000, but translation was better than this. I guess that
I have some problems with my corpora, but can't find out what exactly
it is.
And also while training I got this message, "nohup: ignoring input and
redirecting stderr to stdout"
this was my command:
"nohup nice ~/mosesdecoder/scripts/training/train-model.perl -root-dir
train -corpus ~/corpus/my_corpus -f fr -e en -alignment
grow-diag-final-and -reordering msd-bidirectional-fe -lm
0:6:$HOME/lm/my_language_model:8 -mgiza -external-bin-dir
~/mosesdecoder/tools > training.out "
Could you, please, tell me if you have any idea about this problem?

thanks in advance !

Sehrob
_______________________________________________
Moses-support mailing list
Moses-support@mit.edu
http://mailman.mit.edu/mailman/listinfo/moses-support



------------------------------

_______________________________________________
Moses-support mailing list
Moses-support@mit.edu
http://mailman.mit.edu/mailman/listinfo/moses-support


End of Moses-support Digest, Vol 94, Issue 23
*********************************************

Tweet