<div dir="ltr">Hi,<div><br></div><div>I'm Abu Zaher, a mentor in GSoC2013 from project Apertium and NESCent. I'm one of the developers of en-bn language pair for Apertium which has been in development since 2009 as a GSoC project and further developed in 2011 by another student. This is a large scale project that deals with all the major parts of rule based MT e.g. monolingual dictionary creation (morphological analyser), bilingual dictionary creation, parts of speech tagger training and transfer system design (creation of translation rules). We are looking for prospective students this year for working on this idea as well.</div>
<div><br></div><div>I've come to notice that your org has an idea titled "Add language grammar rules to a machine translation system"[1]. It has a line </div><div><br></div><div><span style="color:rgb(51,51,51);font-family:Helvetica,arial,freesans,clean,sans-serif;font-size:14px;line-height:22px"><i>Existing Machine Translation systems perform with less accuracy when provided with the task of translating from English to Bengali. T</i></span><span style="color:rgb(51,51,51);font-family:Helvetica,arial,freesans,clean,sans-serif;font-size:14px;line-height:22px"><i>he proposal involves documenting existing language grammar rules and developing enhancements to an existing system of machine translation.</i></span><i><br clear="all">
</i><div><br></div><div><div style>I'd like to know which platforms you guys are considering as existing machine translation system and thinking of improving. Have you considered our platform for this idea? To our knowledge Apertium has the largest open source resources dedicated to Bengali-English rule based MT systems. We'd be more than happy to give you a helping hand in the idea if you want. Apertium has a broad range of expertise specially dedicated to MT and we think that could help you in this idea a lot. We think any kind of collaboration (if you want to avail it) would be beneficial for both our organisations.</div>
</div><div style><br></div><div style>I've also added Francis Tyers (the org admin of Apertium, an expert in MT and my mentor from 2009 project) to this thread who can give you a more details if you want. You can find the codebase from our project here[2][3]. Our wiki dedicated to bn-en pair can be accessed from here[4] (it's a dev wiki so it's kinda shabby though).</div>
<div style><br></div><div style>[1] <a href="https://github.com/ankur-india/ankur-india/wiki/Project-Ideas#add-language-grammar-rules-to-a-machine-translation-system">https://github.com/ankur-india/ankur-india/wiki/Project-Ideas#add-language-grammar-rules-to-a-machine-translation-system</a></div>
<div style>[2] <a href="https://apertium.svn.sourceforge.net/svnroot/apertium/incubator/apertium-bn-en/">https://apertium.svn.sourceforge.net/svnroot/apertium/incubator/apertium-bn-en/</a> (svn)</div><div style>[3] <a href="https://github.com/darthxaher/apertium-bn-en">https://github.com/darthxaher/apertium-bn-en</a> (github mirror)</div>
<div style>[4] <a href="http://wiki.apertium.org/wiki/Bengali_and_English">http://wiki.apertium.org/wiki/Bengali_and_English</a></div><div><br></div>-- <br>Regards<br>Abu Zaher Md. Faridee<br><br><a href="http://zaher14.blogspot.com/" target="_blank">http://zaher14.blogspot.com/</a><br>
<a href="http://sourceforge.net/projects/apertium/" target="_blank">http://sourceforge.net/projects/apertium/</a><br>
<div><a href="http://www.nescent.org/">http://www.nescent.org/</a><br></div></div></div>