apertium-stuff@lists.sourceforge.net
To subscribe or unsubscribe via the World Wide Web, visit
https://lists.sourceforge.net/lists/listinfo/apertium-stuff
or, via email, send a message with subject or body 'help' to
apertium-stuff-request@lists.sourceforge.net
You can reach the person managing the list at
apertium-stuff-owner@lists.sourceforge.net
When replying, please edit your Subject line so it is more specific
than "Re: Contents of Apertium-stuff digest..."
Today's Topics:
1. Re: Apertium-stuff Digest, Vol 23, Issue 9 (rushikesh deshpande)
2. Re: Apertium-stuff Digest, Vol 23, Issue 9 (Jimmy O'Regan)
3. Re: n-Stage transfer (Francis Tyers)
4. Gsoc idea multi engine synthesiser (amit sethi)
5. GSoC Idea: Conversion of Anubadok (Abu Zaher)
----------------------------------------------------------------------
Message: 1
Date: Mon, 23 Mar 2009 07:37:57 +0530
From: rushikesh deshpande <rushikesh90@gmail.com>
Subject: Re: [Apertium-stuff] Apertium-stuff Digest, Vol 23, Issue 9
To: apertium-stuff@lists.sourceforge.net
Message-ID:
<3f9904880903221907l3c720b41ocb840e2105287868@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
hi I am computer science student from India.I have a good experience
in c++ language coding.I
would like to take part in developement of n-s tage transfer
idea.Please let me know where i can get the source code.
On 3/23/09, apertium-stuff-request@lists.sourceforge.net
<apertium-stuff-request@lists.sourceforge.net> wrote:
> Send Apertium-stuff mailing list submissions to
> apertium-stuff@lists.sourceforge.net
>
> To subscribe or unsubscribe via the World Wide Web, visit
> https://lists.sourceforge.net/lists/listinfo/apertium-stuff
> or, via email, send a message with subject or body 'help' to
> apertium-stuff-request@lists.sourceforge.net
>
> You can reach the person managing the list at
> apertium-stuff-owner@lists.sourceforge.net
>
> When replying, please edit your Subject line so it is more specific
> than "Re: Contents of Apertium-stuff digest..."
>
>
> Today's Topics:
>
> 1. Re: New Apertium Tool Project (Mikel L. Forcada)
> 2. Re: New Apertium Tool Project (Marcin Mi?kowski)
> 3. A proposal (Ankuj Gupta)
> 4. Re: A proposal (Francis Tyers)
> 5. Re: A proposal (Ankuj Gupta)
> 6. Re: A proposal (Jimmy O'Regan)
> 7. Re: A proposal (Ankuj Gupta)
>
>
> ----------------------------------------------------------------------
>
> Message: 1
> Date: Sun, 22 Mar 2009 17:12:31 +0100
> From: "Mikel L. Forcada" <mlf@dlsi.ua.es>
> Subject: Re: [Apertium-stuff] New Apertium Tool Project
> To: apertium-stuff@lists.sourceforge.net
> Message-ID: <200903221712.31356.mlf@dlsi.ua.es>
> Content-Type: text/plain; charset="utf-8"
>
> Dear all:
>
> Jimmy O'Regan says:
>
>> Correct me if I'm wrong, but didn't we have dixtools in Java before
>> Sun GPLd it? I don't see how the situation is different.
>
> Here's the data on the use of Java at Apertium:
>
> The Java platform was released under the GPL on november 14 ,2006:
> http://www.fsf.org/news/fsf-welcomes-gpl-java.html
>
> Mon May 14 15:12:40 2007 UTC (22 months, 1 week ago): crossdics (old name of
> apertium-dixtools) added to SVN. Can't find it in the old CVS, so that's
> probably it.
>
> Maybe some components used in apertium-dixtools were not free at the time. I
> think Enrique Benimeli can answer this question in more detail. If something
> non-free was done then, I wasn't aware of it.
>
> By the way, I said I personally wasn't going to support applications which
> needed a proprietary player. That is my personal point of view, hopefully
> shared by many other people at Apertium.
>
> In the unlikely case that a qualified majority of people in Apertium decided
> adopting a different philosophy and moving towards generating code that
> generated code that needed proprietary players to work properly (quite
> unlikely, I hope), I would immediately leave Apertium, grab the "good" code,
> and promote a free fork of Apertium somewhere else.
>
> Mikel L. Forcada
>
> --
> Mikel L. Forcada <mlf@dlsi.ua.es>
> http://www.dlsi.ua.es/~mlf/
>
>
>
>
> ------------------------------
>
> Message: 2
> Date: Sun, 22 Mar 2009 17:34:34 +0100
> From: Marcin Mi?kowski <list-address@wp.pl>
> Subject: Re: [Apertium-stuff] New Apertium Tool Project
> To: apertium-stuff@lists.sourceforge.net
> Message-ID: <49C6689A.5000104@wp.pl>
> Content-Type: text/plain; charset=UTF-8; format=flowed
>
> Mikel L. Forcada pisze:
>> Dear all:
>>
>> Jimmy O'Regan says:
>>
>>
>>> Correct me if I'm wrong, but didn't we have dixtools in Java before
>>> Sun GPLd it? I don't see how the situation is different.
>>>
>>
>> Here's the data on the use of Java at Apertium:
>>
>> The Java platform was released under the GPL on november 14 ,2006:
>> http://www.fsf.org/news/fsf-welcomes-gpl-java.html
>>
>> Mon May 14 15:12:40 2007 UTC (22 months, 1 week ago): crossdics (old name
>> of
>> apertium-dixtools) added to SVN. Can't find it in the old CVS, so that's
>> probably it.
>>
>> Maybe some components used in apertium-dixtools were not free at the time.
>> I
>> think Enrique Benimeli can answer this question in more detail. If
>> something
>> non-free was done then, I wasn't aware of it.
>>
> It couldn't have been as Sun Java is not the only Java around, and GCJ
> could have been used as well. As far as I understand, the open-source
> Flash players are much less compatible with the Adobe Flash than GCJ
> with Sun Java, but if the compatibility with them is retained, then the
> whole environment could be kept completely free (free as Stallman
> defines it).
>
> Regards
> Marcin
>
>
>
> ------------------------------
>
> Message: 3
> Date: Mon, 23 Mar 2009 00:43:37 +0530
> From: Ankuj Gupta <ankuj2004@gmail.com>
> Subject: [Apertium-stuff] A proposal
> To: apertium-stuff@lists.sourceforge.net
> Message-ID:
> <a69efbe60903221213v5516dfadh7bf92190a829db92@mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> Hi!!
> I am Computer Engineering Student from India. I had worked on creating a
> bilingual dictionary based on Wikipedia which could be easily used easily to
> translate words which are in far away related languages like for eg English
> to japanese and english to hindi. As of now what i see is that Apertium does
> not Indian languages. So, adding them would give it a global outlook.
> My query is is it possible to add language pairs that it does not support at
> the moment. If I apply this idea as proposal will I be awarded a mentor.
> I would like to contribute to apertium . Since < i have worked on
> translation aspect so i know some of the ascpects gone into online
> translator.
> --
> Ankuj Gupta
> Computer Engineering
> NSIT
> India
> -------------- next part --------------
> An HTML attachment was scrubbed...
>
> ------------------------------
>
> Message: 4
> Date: Sun, 22 Mar 2009 20:21:59 +0100
> From: Francis Tyers <ftyers@prompsit.com>
> Subject: Re: [Apertium-stuff] A proposal
> To: apertium-stuff@lists.sourceforge.net
> Message-ID: <1237749719.3355.4.camel@eki.prompsit.com>
> Content-Type: text/plain; charset=utf-8
>
> El lun, 23-03-2009 a las 00:43 +0530, Ankuj Gupta escribi?:
>> Hi!!
>> I am Computer Engineering Student from India. I had worked on creating
>> a bilingual dictionary based on Wikipedia which could be easily used
>> easily to translate words which are in far away related languages like
>> for eg English to japanese and english to hindi.
>
> Using interwiki links or article text?
>
>> As of now what i see is that Apertium does not Indian languages. So,
>> adding them would give it a global outlook.
>
> It would be good to add them indeed. We have a morphological analyser
> for Hindi already, but it needs to be converted to UTF-8, we have
> various other bits-and-bobs that you can find in the 'incubator' module
> of our SVN.
>
>> My query is is it possible to add language pairs that it does not
>> support at the moment. If I apply this idea as proposal will I be
>> awarded a mentor.
>
> How do you plan on making a new language pair? Do you have experience
> with working with Apertium? Normally making a new language pair takes
> around 6 months, how do you plan to accomplish the task in 3 months?
>
> Best regards,
>
> Fran
>
>
>
>
> ------------------------------
>
> Message: 5
> Date: Mon, 23 Mar 2009 00:59:05 +0530
> From: Ankuj Gupta <ankuj2004@gmail.com>
> Subject: Re: [Apertium-stuff] A proposal
> To: ftyers@prompsit.com, apertium-stuff@lists.sourceforge.net
> Message-ID:
> <a69efbe60903221229x5ae322aco1ee9c87762d27333@mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> I had used interwiki along with redirects and linktexts. I have no prior
> knowledge of apertium . But that can be taken care of if i give my input to
> it. Rather than adding a new pair I can work on the Hindi language part that
> is already existing, will be exploring it. But this opportunity really
> excites me.
>
> On Mon, Mar 23, 2009 at 12:51 AM, Francis Tyers <ftyers@prompsit.com> wrote:
>
>> El lun, 23-03-2009 a las 00:43 +0530, Ankuj Gupta escribi?:
>> > Hi!!
>> > I am Computer Engineering Student from India. I had worked on creating
>> > a bilingual dictionary based on Wikipedia which could be easily used
>> > easily to translate words which are in far away related languages like
>> > for eg English to japanese and english to hindi.
>>
>> Using interwiki links or article text?
>>
>> > As of now what i see is that Apertium does not Indian languages. So,
>> > adding them would give it a global outlook.
>>
>> It would be good to add them indeed. We have a morphological analyser
>> for Hindi already, but it needs to be converted to UTF-8, we have
>> various other bits-and-bobs that you can find in the 'incubator' module
>> of our SVN.
>>
>> > My query is is it possible to add language pairs that it does not
>> > support at the moment. If I apply this idea as proposal will I be
>> > awarded a mentor.
>>
>> How do you plan on making a new language pair? Do you have experience
>> with working with Apertium? Normally making a new language pair takes
>> around 6 months, how do you plan to accomplish the task in 3 months?
>>
>> Best regards,
>>
>> Fran
>>
>>
>>
>> ------------------------------------------------------------------------------
>> Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are
>> powering Web 2.0 with engaging, cross-platform capabilities. Quickly and
>> easily build your RIAs with Flex Builder, the Eclipse(TM)based development
>> software that enables intelligent coding and step-through debugging.
>> Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com
>> _______________________________________________
>> Apertium-stuff mailing list
>> Apertium-stuff@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/apertium-stuff
>>
>
>
>
> --
> Ankuj
> -------------- next part --------------
> An HTML attachment was scrubbed...
>
> ------------------------------
>
> Message: 6
> Date: Sun, 22 Mar 2009 19:39:18 +0000
> From: "Jimmy O'Regan" <joregan@gmail.com>
> Subject: Re: [Apertium-stuff] A proposal
> To: apertium-stuff@lists.sourceforge.net
> Message-ID:
> <e94dc08d0903221239h433c49acqf097527ed4a19674@mail.gmail.com>
> Content-Type: text/plain; charset=UTF-8
>
> 2009/3/22 Ankuj Gupta <ankuj2004@gmail.com>:
>> I had used interwiki along with redirects and linktexts. I have no prior
>> knowledge of apertium . But that can be taken care of if i give my input
>> to
>> it. Rather than adding a new pair I can work on the Hindi language part
>> that
>> is already existing, will be exploring it. But this opportunity really
>> excites me.
>>
>
> Well, there is an open source English-Hindi dictionary
> (http://ltrc.iiit.net/onlineServices/Dictionaries/Shabdanjali_Dwnld.html)
> that can be reused. I would advise you to also look into the ideas
> 'Linguistically-driven filtering of the bilingual phrases used to
> infer shallow-transfer rules' and 'Use of context-dependent
> lexicalized categories in the inference of shallow-transfer rules'
> from our projects page
> (http://wiki.apertium.org/wiki/Ideas_for_Google_Summer_of_Code) to
> help speed up the creation of transfer rules. (It shouldn't be too
> hard to find enough parallel text in English and Hindi)
>
>
>
> ------------------------------
>
> Message: 7
> Date: Mon, 23 Mar 2009 01:19:11 +0530
> From: Ankuj Gupta <ankuj2004@gmail.com>
> Subject: Re: [Apertium-stuff] A proposal
> To: apertium-stuff@lists.sourceforge.net
> Message-ID:
> <a69efbe60903221249k65e8736o8e4a0b0417c14223@mail.gmail.com>
> Content-Type: text/plain; charset="iso-8859-1"
>
> One of the difficulty faced during our work was parallel corpora thats why
> we resorted to Wikipedia. I guess that can work as a source of parallel
> text. Correct me if i am wrong.
>
> On Mon, Mar 23, 2009 at 1:09 AM, Jimmy O'Regan <joregan@gmail.com> wrote:
>
>> 2009/3/22 Ankuj Gupta <ankuj2004@gmail.com>:
>> > I had used interwiki along with redirects and linktexts. I have no prior
>> > knowledge of apertium . But that can be taken care of if i give my input
>> to
>> > it. Rather than adding a new pair I can work on the Hindi language part
>> that
>> > is already existing, will be exploring it. But this opportunity really
>> > excites me.
>> >
>>
>> Well, there is an open source English-Hindi dictionary
>> (http://ltrc.iiit.net/onlineServices/Dictionaries/Shabdanjali_Dwnld.html)
>> that can be reused. I would advise you to also look into the ideas
>> 'Linguistically-driven filtering of the bilingual phrases used to
>> infer shallow-transfer rules' and 'Use of context-dependent
>> lexicalized categories in the inference of shallow-transfer rules'
>> from our projects page
>> (http://wiki.apertium.org/wiki/Ideas_for_Google_Summer_of_Code) to
>> help speed up the creation of transfer rules. (It shouldn't be too
>> hard to find enough parallel text in English and Hindi)
>>
>>
>> ------------------------------------------------------------------------------
>> Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are
>> powering Web 2.0 with engaging, cross-platform capabilities. Quickly and
>> easily build your RIAs with Flex Builder, the Eclipse(TM)based development
>> software that enables intelligent coding and step-through debugging.
>> Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com
>> _______________________________________________
>> Apertium-stuff mailing list
>> Apertium-stuff@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/apertium-stuff
>>
>
>
>
> --
> Ankuj
> -------------- next part --------------
> An HTML attachment was scrubbed...
>
> ------------------------------
>
> ------------------------------------------------------------------------------
> Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are
> powering Web 2.0 with engaging, cross-platform capabilities. Quickly and
> easily build your RIAs with Flex Builder, the Eclipse(TM)based development
> software that enables intelligent coding and step-through debugging.
> Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com
>
> ------------------------------
>
> _______________________________________________
> Apertium-stuff mailing list
> Apertium-stuff@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/apertium-stuff
>
>
> End of Apertium-stuff Digest, Vol 23, Issue 9
> *********************************************
>
--
rushikesh d d
------------------------------
Message: 2
Date: Mon, 23 Mar 2009 02:34:25 +0000
From: "Jimmy O'Regan" <joregan@gmail.com>
Subject: Re: [Apertium-stuff] Apertium-stuff Digest, Vol 23, Issue 9
To: apertium-stuff@lists.sourceforge.net
Message-ID:
<e94dc08d0903221934t4d076d09med5245515aa4f668@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8
2009/3/23 rushikesh deshpande <rushikesh90@gmail.com>:
> hi I am computer science student from India.I have a good experience
> in c++ language coding.I
> would like to take part in developement of n-s tage transfer
> idea.Please let me know where i can get the source code.
>
SVN details are here: http://sourceforge.net/scm/?type=svn&group_id=143781
You'll be working with the module 'apertium' (trunk/apertium);
specifically, with the interchunk* files.
------------------------------
Message: 3
Date: Mon, 23 Mar 2009 07:52:48 +0100
From: Francis Tyers <ftyers@prompsit.com>
Subject: Re: [Apertium-stuff] n-Stage transfer
To: apertium-stuff@lists.sourceforge.net
Message-ID: <1237791168.3355.46.camel@eki.prompsit.com>
Content-Type: text/plain; charset=UTF-8
El lun, 23-03-2009 a las 07:37 +0530, rushikesh deshpande escribi?:
> hi I am computer science student from India.I have a good experience
> in c++ language coding.I
> would like to take part in developement of n-s tage transfer
> idea.Please let me know where i can get the source code.
Please feel free to document your ideas here:
http://wiki.apertium.org/wiki/N-Stage_transfer
Fran
------------------------------
Message: 4
Date: Mon, 23 Mar 2009 12:26:49 +0530
From: amit sethi <amit.pureenergy@gmail.com>
Subject: [Apertium-stuff] Gsoc idea multi engine synthesiser
To: apertium-stuff@lists.sourceforge.net
Message-ID:
<da81a0a80903222356h687b92dbt2ef7334977a261cb@mail.gmail.com>
Content-Type: text/plain; charset="iso-8859-1"
I am interested in applying for multi-engine language synthesizer :
http://wiki.apertium.org/wiki/Ideas_for_Google_Summer_of_Code
The idea uses the open translation Engines available to synthesize a better
translation.
I propose that this can be done with vector space
model<http://en.wikipedia.org/wiki/Vector_space_model>.
Steps:
Break text into pieces that can be deemed comparable.
The broken texts are compared .The vector space model gives a similarity
score to each translation engine pair and the one which has the highest
scoring(i.e the one that gives a result that all engines agree with) is the
most trustable engine . I am not sure if other such dynamic scoring system
exist which could help us in this task but would love any suggestions on
already existing pieces.
I am interested in developing this project but not sure of algorithms or
tools that could help us accomplish this task.
I have worked on language processing a bit and have an idea of concepts like
stemming, inverse document frequency etc that might be useful.
I am also thus not sure of quantum of work that might be required for the
project.
And would love suggestion from the list.
--
A-M-I-T S|S
-------------- next part --------------
An HTML attachment was scrubbed...
------------------------------
Message: 5
Date: Mon, 23 Mar 2009 13:37:57 +0600
From: Abu Zaher <zaher14@gmail.com>
Subject: [Apertium-stuff] GSoC Idea: Conversion of Anubadok
To: apertium-stuff@lists.sourceforge.net
Message-ID:
<2de11de20903230037x3efaaa7br13c579119b02cf52@mail.gmail.com>
Content-Type: text/plain; charset="utf-8"
Hi,
I'm an undergrad student of Computer Science from Bangladesh and currently
attending my 3rd year. I'm interested in working on "*Conversion of Anubadok
*". I have work experience in C/C++/flex/yacc as well as perl/python. As
Bengali is my first language, I'd like to take this opportunity to better
serve my nation. I'm also panning to work further on Bengali-English
transtion as part of my undergrad thesis which could benefit from this.
I've alreay contacted the author of Anubadok Mr. G.M. Hossain and he has
given me some insight on Anubadok. From what I've seen in such little time,
there is much work needs to be done in the Bengali generaion side. Notably,
1. Currently it supports Simple and Compound sentence translation but lacks
support for Complex and Complex-Compound sentences.
2. Anubadok has difficulty choosing between multiple meaning of the same
word depending on the contex.
3. Also the Parts of Speech tagger from Anubadok is deffirent from the
tagger from Apertium.
It seems a daunting task to get all this done in the allocated time but I'm
sure I can pull it off.
Please feel free to share your insight/implications/viabiliy on this
particular Idea for a GSoC project.
--
Regards
*Abu Zaher Md. Faridee*
Department of Computer Science and Engineering
Bangladesh University of Engineering and Technology
http://zaher14.blogspot.com/
---
Time heals every wound, but time itself is a wound that never heals.
-------------- next part --------------
An HTML attachment was scrubbed...
------------------------------
------------------------------------------------------------------------------
Apps built with the Adobe(R) Flex(R) framework and Flex Builder(TM) are
powering Web 2.0 with engaging, cross-platform capabilities. Quickly and
easily build your RIAs with Flex Builder, the Eclipse(TM)based development
software that enables intelligent coding and step-through debugging.
Download the free 60 day trial. http://p.sf.net/sfu/www-adobe-com
------------------------------
_______________________________________________
Apertium-stuff mailing list
Apertium-stuff@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/apertium-stuff
End of Apertium-stuff Digest, Vol 23, Issue 11
**********************************************
No comments:
Post a Comment