Difference between revisions of "Mirandese and Portuguese"
From LING073
(→Final Evaluation) |
|||
| (3 intermediate revisions by 2 users not shown) | |||
| Line 37: | Line 37: | ||
**Proportion of stems translated correctly in the longer corpus: 315 | **Proportion of stems translated correctly in the longer corpus: 315 | ||
**Trimmed coverage over longer and large corpora: | **Trimmed coverage over longer and large corpora: | ||
| + | ***Longer: 421 / 596 (~0.70637583892617449664) | ||
| + | ***Large: 1574090 / 3075136 (~0.51187654789901975067) | ||
**Number of tokens in longer and large corpora: | **Number of tokens in longer and large corpora: | ||
| + | ***Longer: 608 | ||
| + | ***Large: 3078793 | ||
[[Category:Sp23_TranslationPairs]] [[Category:Mirandese]] [[Category:Portuguese]] | [[Category:Sp23_TranslationPairs]] [[Category:Mirandese]] [[Category:Portuguese]] | ||
Latest revision as of 18:48, 28 April 2023
Resources for machine translation between Mirandese and Portuguese
Developed Resources
- Corpus Repo
- Mirandese and Portuguese/Evaluation
- Mirandese and Portuguese/Lexical selection
- Mirandese and Portuguese/Contrastive Grammar
- Mirandese and Portuguese/Structural transfer
External Resources
Additions
- 5 mwl paradigms
- 3 mwl disambiguation rules
- 3 por disambiguation rules
Final Evaluation
- Monolingual Transducer:
- Updated precision & recall:
- Precision: 98.26590%
- Recall: 57.23906%
- Coverage over large corpus: ~0.51365440837854797405
- Number of words in large corpus: 3066116
- Number of stems in the transducer: 426
- Updated precision & recall:
- mwl-por:
- WER and PER over large corpus:
- WER: 37.82 %
- PER: 34.24 %
- Proportion of stems translated correctly in the longer corpus: 315
- Trimmed coverage over longer and large corpora:
- Longer: 421 / 596 (~0.70637583892617449664)
- Large: 1574090 / 3075136 (~0.51187654789901975067)
- Number of tokens in longer and large corpora:
- Longer: 608
- Large: 3078793
- WER and PER over large corpus:
