Difference between revisions of "User:Deltamachine/proposal"
Deltamachine (talk | contribs) |
Deltamachine (talk | contribs) |
||
| Line 57: | Line 57: | ||
=== Work period === |
=== Work period === |
||
<ul> |
<ul> |
||
| − | <li>'''1st month:''' preparing the data, proceeding treebanks, creating datasets for training</li> |
+ | <li>'''1st month:''' preparing the data, proceeding treebanks, creating datasets for training.</li> |
| − | <li>'''2nd month: working on a classifier, testing |
+ | <li>'''2nd month:''' working on a classifier, testing.</li> |
| − | <li>'''3rd month: integrating shallow function labeller to Apertium, testing, fixing bugs, writing documentation |
+ | <li>'''3rd month:''' integrating shallow function labeller to Apertium, testing, fixing bugs, writing documentation.</li> |
</ul> |
</ul> |
||
Revision as of 07:02, 15 March 2017
Contents
- 1 Contact information
- 2 Skills and experience
- 3 Why is it you are interested in machine translation?
- 4 Why is it that you are interested in Apertium?
- 5 Which of the published tasks are you interested in? What do you plan to do?
- 6 Reasons why Google and Apertium should sponsor it
- 7 A description of how and who it will benefit in society
- 8 Work plan
- 9 Non-Summer-of-Code plans you have for the Summer
- 10 Coding challenge
Contact information
Name: Anna Kondratjeva
Location: Moscow, Russia
E-mail: an-an-kondratjeva@yandex.ru
Phone number: +79250374221
Github: http://github.com/deltamachine
IRC: deltamachine
Timezone: UTC+3
Skills and experience
Education: Bachelor's Degree in Fundamental and Computational Linguistics (2015 - expected 2019), National Research University «Higher School of Economics» (NRU HSE)
Main university courses:
- Theory of Language (Phonetics, Morphology, Syntax, Semantics)
- Programming (Python)
- Computer Tools for Linguistic Research
- Language Diversity and Typology
- Introduction to Data Analysis
- Math (Discrete Math, Linear Algebra and Calculus, Probability Theory and Mathematical Statistics)
Technical skills: Python (experienced, 1.5 years), HTML, CSS, Flask, Django, SQLite (familiar)
Projects and experience: http://github.com/deltamachine
Languages: Russian (native), English, German
Why is it you are interested in machine translation?
Why is it that you are interested in Apertium?
Which of the published tasks are you interested in? What do you plan to do?
I would like to implement a prototype shallow syntactic function labeller.
Reasons why Google and Apertium should sponsor it
A description of how and who it will benefit in society
Work plan
Post application period
- Getting closer with Apertium, reading documentation, playing around with its tools
- Setting up Linux and getting used to it
- Learning more about UD treebanks
- Learning more about machine learning
Community bonding period
- Choosing language pairs, with which shallow function labeller will work.
- Choosing the most appropriate Python ML library (maybe it will be Tensorflow, maybe not)
Work period
- 1st month: preparing the data, proceeding treebanks, creating datasets for training.
- 2nd month: working on a classifier, testing.
- 3rd month: integrating shallow function labeller to Apertium, testing, fixing bugs, writing documentation.
Schedule
- Week 1:
- Week 2:
- Week 3:
- Week 4:
- Deliverable #1, June 26 - 30:
- Week 5:
- Week 6:
- Week 7:
- Week 8:
- Deliverable #2, July 24 - 28:
- Week 9:
- Week 10:
- Week 11:
- Week 12:
- Project completed
Non-Summer-of-Code plans you have for the Summer
I have exams in the university till 3rd week of June, so I will be able to work only 20-25 hours per week. But I will try to pass as many exams as possible ahead of schedule, so it may be changed. After that I will be able to work full time and spend 45-50 hours per week on the task.
Coding challenge
https://github.com/deltamachine/wannabe_hackerman
- flatten_conllu.py: A script that takes a dependency treebank in UD format and "flattens" it, that is, applies the following transformations:
- Words with the @conj relation take the label of their head
- Words with the @parataxis relation take the label of their head
- calculate_accuracy_index.py: A script that does the following:
- Takes -train.conllu file and calculates the table: surface_form - label - frequency
- Takes -dev.corpus and for each token assigns the most frequent label from the table
- Calculates the accuracy index
- label_asf: A script that takes a sentence in Apertium stream format and for each surface form applies the most frequent label from the labelled corpus.
