Difference between revisions of "Miskito/Final project"
(→Overview) |
|||
| Line 36: | Line 36: | ||
==Evaluation== | ==Evaluation== | ||
| − | |||
| − | |||
To evaluate our progress, we used the metric <code>coverage-hfst</code> to record our overall corpus coverage. Below is our reported results. | To evaluate our progress, we used the metric <code>coverage-hfst</code> to record our overall corpus coverage. Below is our reported results. | ||
| Line 52: | Line 50: | ||
# Adding stems for high-frequency words | # Adding stems for high-frequency words | ||
# Implementing morphological rules (ie. tenses) | # Implementing morphological rules (ie. tenses) | ||
| + | |||
| + | |||
| + | |||
| + | |||
| + | [[Category:sp21_FinalProjects]] [[Category:Miskito]] | ||
Revision as of 13:34, 20 May 2021
Contents
Overview
Over the course of the semester, we've worked on developing a rule based machine translation (RBMT) system for Miskito utilizing apertium resources.
Implementation
For our final project we decided to focus on our monolingual transducer and improve it as much as possible. Specifically, we directed our attention to the .lexd and .twol files to encode grammatical generalizations. We managed to implement a large range of morphological analysis for our language.
- Verb tense morphology
- Absolute Past
- Present
- Present Progressive
- Absolute Future
- Future
- Irregular Verb Morphology
- Negation in the present tense
- Possessive Noun morphology
With just a combined set of around 230 nouns and verbs, we achieved an impressive amount of coverage detailed below.
Areas to fix
- Full coverage of possessive noun morphology
- Adjective morphology
Developed Resources
- Corpus Repo
- Miskito Keyboard
- Grammar Documentation
- Miskito Transducer
- Disambiguation wiki
- Miskito and English Wiki Page
- Miskito to English Apertium Language Pair Repo
- Contrastive Grammars wiki
- Link to the code for our morphological analyzer
We expanded our morphological analyzer's capabilities by adding in
Evaluation
To evaluate our progress, we used the metric coverage-hfst to record our overall corpus coverage. Below is our reported results.
Initial Corpus Coverage
- Coverage: 3873 / 7569 (~0.51169242964724534285)
- Remaining unknown forms: 3696
Final Corpus Coverage
- Coverage: 5987 / 7588 (~0.78900896151818661044)
- Remaining unknown forms: 1601
We were able to attain this increase in overall coverage by using two strategies:
- Adding stems for high-frequency words
- Implementing morphological rules (ie. tenses)
