Jya and English/Structural transfer

Pre-evaluation

Number of words in reference: 20
Number of words in test: 18
Number of unknown words (marked with a star) in test: 4
Percentage of unknown words: 22.22 %

Results when removing unknown-word marks (stars)
-------------------------------------------------------
Edit distance: 16
Word error rate (WER): 80.00 %
Number of position-independent correct words: 5
Position-independent word error rate (PER): 75.00 % 

Implementation / Development

Appropriate adjective-noun order has been implemented:

 $ echo "mbndəd magənndɽa" | apertium -d . jya-eng-lex
2:SELECT:2:magenndra<adj>:3:terrible<adj>
^mbnded<n>/demon<n>$ ^magenndra<adj>/terrible<adj>$ 

Now, the words have the correct order:

$ echo "mbndəd magənndɽa" | apertium -d . jya-eng
terrible demon 

Verb-final word order has also been implemented:

$ echo "rmu gənndza" | apertium -d . jya-eng
eat person 

The full test set now produces no unknowns or generation failures:

$ cat jya.tests.txt | apertium -d . jya-eng
the demon
the king
a king
be
put
terrible demon
terrible king
eat person
be king
his friend
his wife 

Post-evaluation

Number of words in reference: 20
Number of words in test: 20
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: 2
Word error rate (WER): 10.00 %
Number of position-independent correct words: 18
Position-independent word error rate (PER): 10.00 %