Difference between revisions of "Uyghur and English/Structural Transfer"
From LING073
(→Pre-Evaluation) |
|||
| Line 11: | Line 11: | ||
------------------------------------------------------- | ------------------------------------------------------- | ||
| − | Number of words in reference: | + | Number of words in reference: 62 |
| − | Number of words in test: | + | Number of words in test: 46 |
Number of unknown words (marked with a star) in test: 4 | Number of unknown words (marked with a star) in test: 4 | ||
| − | Percentage of unknown words: | + | Percentage of unknown words: 8.70 % |
Results when removing unknown-word marks (stars) | Results when removing unknown-word marks (stars) | ||
| Line 23: | Line 23: | ||
------------------------------------------------------- | ------------------------------------------------------- | ||
| − | Edit distance: | + | Edit distance: 61 |
| − | Word error rate (WER): | + | Word error rate (WER): 98.39 % |
Number of position-independent correct words: 2 | Number of position-independent correct words: 2 | ||
| − | Position-independent word error rate (PER): | + | Position-independent word error rate (PER): 96.77 % |
[[Category:Sp25_StructuralTransfer]] | [[Category:Sp25_StructuralTransfer]] | ||
Revision as of 20:26, 11 May 2025
scrapeTransferTests -p uig-eng "Uyghur_and_English/Contrastive_Grammar"
cat uig.tests.txt | apertium -d ../ling073-uig-eng uig-eng > uig-eng.tests.txt
apertium-eval-translator-line -r eng.tests.txt -t uig-eng.tests.txt
Pre-Evaluation
Statistics about input files
Number of words in reference: 62
Number of words in test: 46
Number of unknown words (marked with a star) in test: 4
Percentage of unknown words: 8.70 %
Results when removing unknown-word marks (stars)
Edit distance: 61
Word error rate (WER): 98.39 %
Number of position-independent correct words: 2
Position-independent word error rate (PER): 96.77 %
