[Student-projects] GSoC 2014 - Student introduction

[Navaneeth K N]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA512

Hello Anirudh,

On 2/27/14 11:21 PM, Anirudh Vemula wrote:
> I have found the instructions to build Varnam (from Gitorius itself). I am
> looking forward to a detailed description of what the project is intended
> to achieve.

I am assuming you are talking about the improve learning system idea.

Varnam has a learning system built-in which can learn words and it can
also learn possible other ways to write a word.

For eg:
	learn("भारत") = [bharat, bhaarath, bharath]
	transliterate("bharat") = भारत
	transliterate("bhaarath") = भारत
	transliterate("bharath") = भारत

Varnam also learns a word's prefixes so that it can produce better
predictions for any word which has the same prefix. So in this case,
with just learning the word "भारत", varnam can predict "bharateey" =
"भारतीय".

The proposed idea talks about making this learn better. One example is
infer the word "भारत" when learning भारतीय. Something like a porter
stemmer implementation but integrated into the varnam framework so that
new language support can be added easily.

In your case since you speak only Telugu, I am not sure how you can pick
this up. Because today, varnam supports only Hindi and Malayalam.
Probably you can add Telugu support and work on this idea.

Let me know if something is unclear.


> 
> Thanks
> Anirudh
> 
> 
> On 27 February 2014 23:18, Anirudh Vemula <anirudhfoss at gmail.com> wrote:
> 
>> Hello all,
>>
>> I am Anirudh Vemula, a 3rd year undergraduate at IIT Bombay, India. My
>> hometown is Hyderabad and I speak Telugu(partially, one of the reasons I am
>> interested in SMC). I am doing my Major in Computer Science and
>> Engineering. My interests, in general, include reading novels, programming
>> and gaming. In programming, my interests broadly include functional
>> programming, graphics programming and game programming. I am planning to
>> pursue higher studies (an MS or PhD) in Machine Learning.
>>
>> I haven't contributed to any open source project before this but I am sure
>> GSoC would serve as the best platform to introduce me to this area. I
>> follow the development of projects such as SimpleCV, Git and PaGMO closely
>> and I am part of their mailing lists as I use their projects extensively. I
>> have had my share of experience in Computer graphics. I have completed a
>> basic ML course and I am currently doing my Advanced ML course at IITB.
>>
>> I have gone through the list of Project ideas put up in the SOC page.
>> Among those projects, I really liked the one which tackled the issue of '*Improving
>> the learning system*' of Varnam. I have played around with the online
>> Varnam editor and also downloaded the source code of Varnam(libvarnam) from
>> Gitorius.
>>
>> The actual improvement that is expected of the student is not mentioned in
>> the description of the project in the page. I would like to know what is
>> expected of the student who takes up this project and what is the current
>> method used by Varnam to suggest the words?
>>
>> Also, can anyone point me to where I can get instructions to build Varnam?
>>
>> Thanks
>> Anirudh
>>
> 
> 
> 
> _______________________________________________
> Student-projects mailing list
> Student-projects at lists.smc.org.in
> http://lists.smc.org.in/listinfo.cgi/student-projects-smc.org.in
> 

- -- 
Cheers,
Navaneeth
-----BEGIN PGP SIGNATURE-----
Version: GnuPG/MacGPG2 v2.0.22 (Darwin)
Comment: GPGTools - https://gpgtools.org

iQEcBAEBCgAGBQJTD38oAAoJEHFACYSL7h6kz4gH/1G32h+nvlvmrhO7vZDRaOIk
IV2KZSEZAuXh1glJn6isMtXKyMpAV2H2tph7dT6l0+dD7GzW862KZjYYDrXXZxA8
9q+nLiUTika730ZFlzWjPuyQVkf9jasPlfwczgk18DdGtTE3pbRAfa1Vvc50XV2v
VBGRguQV/aENdzSyd0o9wlEmstD/O+zpvtpVxcDY7QbgdjC4TVO2MU4EuElnn3rF
Ga1Tu313V+aq9rlKZjJc5SsigPPBoQLnqwDyPyN9Hx+BWZr79LISSn7kG3GS2asy
8pF2RhwHCOeT5mqugzrFBHYlHVjRe2yHlPR+6aikWMsasiP2YJWgdVh6iXZ9U3k=
=PUwu
-----END PGP SIGNATURE-----