Difference between revisions of "User:Tachyons/GSoC Proposal"

From Apertium
Jump to navigation Jump to search
Line 103: Line 103:
|-
|-
| week 10
| week 10
| Testvoc
| Testvoc for noun and adjective




Line 109: Line 109:
|-
|-
| week 11
| week 11
| Testvoc
| Testvoc for verb and adverb
|-
|-

Revision as of 12:58, 16 March 2014

Personal Information

Name: Aboobacker MK
E-mail address: aboobackervyd@gmail.com
Wiki : tachyons
Irc nick : tachyons
GIt  : tachyons

Why is it you are interested in machine translation?

I am computer science student,who is tech enthusiast . when I was chatting with friend he used Google translate for translating contents to his language . But there was no Malayalam in Google translate or other MT tools . so I decide to create one ,But immediately I realized that it is not an easy task as it sounds . I googled about it and got article about MT, I learned basics including automata theory and decided to concentrate more on MT. MT plays an important role in information sharing .Using MT We can even read articles written in other language without learning that language

Why is it that they are interested in the Apertium project?

Because I am a foss enthusiast and interested in machine language translation. I saw the name of Apertium in GSoC organisation list . I quickly went through the wiki and realised that this is the project I was searching for. I introduced myself in mailing list and irc and got positive response and help

Which of the published tasks are you interested in? What do you plan to do?

Apertium Malayalam-English pair

Proposal

Adopt a language pair: mal-eng(Malayalam to English)

Malayalam is the official language of Kerala (an Indian state) , It spoken by 40 million people around the world . But there is no machine language translator available for this language. My proposal is to create an English Malayalam pair using Apertium platform, It's complex task because structure of English and Malayalam So It is impossible to create complete translation system in 3 months. but I can do many things in 3 months which can help future development

why Google and Apertium should sponsor it,

because there is no existing machine language tools available for this language ,apertium have a solid background in the MT field so it will helpful to realize my dream . If Google approved this proposal I can spend more time for I will get more technical support

How and who it will benefit in society

Machine language translation can play an important role in Information technology to common people, It will help quick translation of documents s,news etc to other language(Especially to mother tongue)more easily.

work plan (including, if possible, a brief schedule with milestones and deliverables).

Work plan

week plan
week 1 Complete noun morphology


week 2 Verb morphology
week 3 Complete verb morphology, and start adverb
week 4 complete adverb and adjective
week 5 Complete Adjective and other pos tags
week 6 Add more entries in monodix , Testing and debugging , Documentation
Deliverable # 1 Morphological analyser ,
week 7 Add entries in bilingual dictionary
week 8 Morphological Disambiguation with CG


week 7 Lexical selection rules
Deliverable # 2
week 8 mal-eng transfer rules
week 9 eng-mal transfer rules


week 10 Testvoc for noun and adjective


week 11 Testvoc for verb and adverb
week 12 Cleanup and dissemination


.

List your skills and give evidence of your qualifications. Tell us what is your current field of study, major, etc.

I have been involved with Apertium since 2013. I can code in python,c,c++,c# ,java,php ,javascript and little objective c and Google Go. Malayalam is my mother tongue and English was my second language at school , I am familiar with both languages

Convince us that you can do the work. In particular we would like to know whether you have programmed before in open-source projects.

I am new to the open source world so currently my contributions are very less :-(, Check my github [1] profile for my existing open source projects . I am already familiar with apertium ecosystem , current status of my work is available here [2] . I created a simple gui tool for mal-eng pair

Mal-eng-tools.png

List any non-Summer-of-Code plans you have for the Summer, especially employment, if you are applying for internships, and class-taking. Be specific about schedules and time commitments.

I can spend ~30 hours per week for this task, I am not doing any internships right now .I am a freelance programmer But I will pause it if I got selected for GSoC