Difference between revisions of "Dzongkha and English/Structural Transfer"

From LING073
Jump to: navigation, search
(Tagger)
(Transfer)
Line 63: Line 63:
  
 
==== Transfer ====
 
==== Transfer ====
^the{{tag|det}}{{tag|def}}{{tag|sg}}$ ^cat{{tag|n}}{{tag|m}}{{tag|sg}}$
+
^catch{{tag|vblex}}{{tag|tv}}{{tag|p3}}{{tag|sg}}$ ^the{{tag|det}}{{tag|def}}{{tag|sg}}$ ^rat{{tag|n}}{{tag|m}}{{tag|sg}}$
  
 
==== Lex ====
 
==== Lex ====

Revision as of 03:09, 25 May 2021

Pre-Evaluation

Statistics about input file

  • Number of words in reference: 31
  • Number of words in test: 10
  • Number of unknown words (marked with a star) in test: 0
  • Percentage of unknown words: 0.00 %

Results when removing unknown-word marks (stars)

  • Edit distance: 31
  • Word error rate (WER): 100.00 %
  • Number of position-independent correct words: 0
  • Position-independent word error rate (PER): 100.00 %

Results when unknown-word marks (stars) are not removed

  • Edit distance: 31
  • Word Error Rate (WER): 100.00 %
  • Number of position-independent correct words: 0
  • Position-independent word error rate (PER): 100.00 %

Statistics about the translation of unknown words

  • Number of unknown words which were free rides: 0
  • Percentage of unknown words that were free rides: 0%

Coverage

  • Bilingual Coverage: 100%

Example

Determiner After Noun

Dzo-Eng

  • (dzo) བྱི་ལི་དེ་ → (eng) the cat

Tagger

  • (dzo) བྱི་ལི་<n> དེ་<det><def> → (eng) the<det><def> cat<n><sg>

Biltrans

^བྱི་ལི་<n>/cat<n>$^དེ་<det>/the<det><def>$

Transfer

^the<det><def><sg>$ ^cat<n><m><sg>$

Lex

^default<NP><ND>{

^བྱི་ལི་<n>/cat<n>$
^དེ་<det>/the<det><def>$

}$

Verb-Noun Order

Dzo-Eng

  • (dzo) བྱི་ཙི་དེ་ཟུང་ → (eng) catch the rat

Tagger

  • (dzo) བྱི་ཙི་<n> དེ་<det><def> ཟུང་<v><tv> → (eng) catch<vblex> the<det><def> cat<n><sg>

Biltrans

^བྱི་ཙི་<n>/rat<n>$^དེ་<det>/the<det><def>$^ཟུང་<v><tv>/catch<vblex>$

Transfer

^catch<vblex><tv><p3><sg>$ ^the<det><def><sg>$ ^rat<n><m><sg>$

Lex

^default<NP><ND>{

^བྱི་ལི་<n>/cat<n>$
^དེ་<det>/the<det><def>$

}$

Post-Evaluation

Statistics about input files

  • Number of words in reference: 31
  • Number of words in test: 13
  • Number of unknown words (marked with a star) in test:
  • Percentage of unknown words: 0.00 %

Results when removing unknown-word marks (stars)

  • Edit distance: 28
  • Word error rate (WER): 90.32 %
  • Number of position-independent correct words: 3
  • Position-independent word error rate (PER): 90.32 %

Results when unknown-word marks (stars) are not removed

  • Edit distance: 28
  • Word Error Rate (WER): 90.32 %
  • Number of position-independent correct words: 3
  • Position-independent word error rate (PER): 90.32 %

Statistics about the translation of unknown words

  • Number of unknown words which were free rides: 0
  • Percentage of unknown words that were free rides: 0%

Coverage

  • Bilingual Coverage: 100%