Advertisement
Guest User

Untitled

a guest
Apr 19th, 2013
59
0
Never
Not a member of Pastebin yet? Sign Up, it unlocks many cool features!
text 1.09 KB | None | 0 0
  1. Test file: 'paper/evaluation/kaz-tat.19042013.txt'
  2. Reference file 'paper/evaluation/kaz-tat.postedit.txt'
  3.  
  4. Statistics about input files
  5. -------------------------------------------------------
  6. Number of words in reference: 2934
  7. Number of words in test: 2937
  8. Number of unknown words (marked with a star) in test: 43
  9. Percentage of unknown words: 1.46 %
  10.  
  11. Results when removing unknown-word marks (stars)
  12. -------------------------------------------------------
  13. Edit distance: 1033
  14. Word error rate (WER): 35.21 %
  15. Number of position-independent correct words: 1996
  16. Position-independent word error rate (PER): 32.07 %
  17.  
  18. Results when unknown-word marks (stars) are not removed
  19. -------------------------------------------------------
  20. Edit distance: 1076
  21. Word Error Rate (WER): 36.67 %
  22. Number of position-independent correct words: 1953
  23. Position-independent word error rate (PER): 33.54 %
  24.  
  25. Statistics about the translation of unknown words
  26. -------------------------------------------------------
  27. Number of unknown words which were free rides: 43
  28. Percentage of unknown words that were free rides: 100.00 %
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement