Difference between revisions of "Munsee/Final Project"
From LING073
(→Approach) |
(→Approach) |
||
| Line 5: | Line 5: | ||
# We first generated a parallel corpus of verb phrases in Munsee and English. | # We first generated a parallel corpus of verb phrases in Munsee and English. | ||
# We added verb stems to our bilingual dictionary (.dix) and monolingual transducer (.lexd), creating twol rules as necessary. | # We added verb stems to our bilingual dictionary (.dix) and monolingual transducer (.lexd), creating twol rules as necessary. | ||
| + | ## New twol rules for VTIs that take nouns, .... | ||
# We then created syntax trees to model the breakdown of various verb phrases. | # We then created syntax trees to model the breakdown of various verb phrases. | ||
## Structural transfer rules were developed by tracking tags backwards up the trees | ## Structural transfer rules were developed by tracking tags backwards up the trees | ||
| Line 15: | Line 16: | ||
### Subject | ### Subject | ||
# We implemented structural transfer rules for translation from Munsee to English. | # We implemented structural transfer rules for translation from Munsee to English. | ||
| + | ## Issues resolved after presentation on 5/9/25: | ||
| + | ### Implemented "it" objects (applicable to transitive verbs) | ||
| + | ### Implemented subject tracking for transitive verbs | ||
# Finally, we evaluated our output (see evaluation section below) | # Finally, we evaluated our output (see evaluation section below) | ||
Revision as of 23:55, 13 May 2025
Project Focus
Despite ongoing efforts to revitalize the Munsee language, relatively few pedagogical resources are readily available to the public. Our tool aims to support a crucial step in the learning of this language: the translation of the grammatically complex verbs in Munsee to English, a language likely to be spoken by learners of Munsee.
Approach
- We first generated a parallel corpus of verb phrases in Munsee and English.
- We added verb stems to our bilingual dictionary (.dix) and monolingual transducer (.lexd), creating twol rules as necessary.
- New twol rules for VTIs that take nouns, ....
- We then created syntax trees to model the breakdown of various verb phrases.
- Structural transfer rules were developed by tracking tags backwards up the trees
- Attributes currently being tracked:
- Transitivity
- Animacy
- Number
- Polarity
- Gender
- Subject
- We implemented structural transfer rules for translation from Munsee to English.
- Issues resolved after presentation on 5/9/25:
- Implemented "it" objects (applicable to transitive verbs)
- Implemented subject tracking for transitive verbs
- Issues resolved after presentation on 5/9/25:
- Finally, we evaluated our output (see evaluation section below)
Outstanding Tasks
- Develop structural transfer rules so that objects of verbs can be tracked and translated
- This will allow our tool to handle transitive verbs
- Account for differences between stable and unstable verb stems
- This will require work on both .lexd and .twol files
- Implement future 'tense' with modal word 'uch'
Relevant Links
- Munsee Corpus Repository
- Munsee Transducer Repository
- Repository for English transducer
- Repository for Munsee-English language pair
Evaluation
As of May 8, 2025
Line Evaluation Results
- Number of words in reference: 175
- Number of words in test: 82
- Number of unknown words in test: 50
- Percentage of unknown words: 60.98%
- Word Error Rate (WER): 97.14%
- Position-Independent Word Error Rate (PER): 97.14%
