[Project-ideas] Fw: Add language grammar rules to a machine translation system

[Sankarshan Mukhopadhyay]

On Fri, Mar 30, 2012 at 11:29 PM, Turja Kundu <turja_cse06 at yahoo.com> wrote:

> as your recommendation i look an existing open source MT system named
> Anubadok: The Bengali Machine Translator. I read the documenatation of how
> it works and also go through the Anubadok's SVN repository. By going through
> the Anubadok libary and reading its documentation  i also have the basic
> idea how it works. From the translator's current status table i learn that
> simple sentences of Bangla is well implemented,compounds are moderately
> implemented, complex and complex-compound sentences are not
> well-implemented.So there is a chance to work on there.

Have you had a look at
<http://wiki.apertium.org/wiki/Google_Summer_of_Code/Report_2009#Conversion_of_Anubadok_.28darthxaher.29>

> i dont understand that have u any plan or roadmap to enrich it's
> grammertical rules,i work along the roadmap or i  make my proposal on my own
> way.

You are expected to come up with a proposal based on your
understanding of the problem and convince the mentor(s) that it is
indeed the optimal way.


-- 
sankarshan mukhopadhyay
<http://sankarshan.randomink.org/blog/>