Japhug and English/Structural Transfer
Contents
Pre-Evaluation
Statistics About Input Files
Number of words in reference: 47 Number of words in test: 33 Number of unknown words (marked with a star) in test: 18 Percentage of unknown words: 54.55 %
Results when removing unknown-word marks (stars)
Edit distance: 47 Word error rate (WER): 100.00 % Number of position-independent correct words: 0 Position-independent word error rate (PER): 100.00 %
Coverage Test on jya.tests.txt in Bilingual Transducer
TOP UNKNOWN WORDS:
1 ^ʁnɯtɤrtsɯɣ/*ʁnɯtɤrtsɯɣ$
1 ^ɲɯrganɯ/*ɲɯrganɯ$
1 ^ɲɤmtsɯr/*ɲɤmtsɯr$
1 ^ɯʑo/*ɯʑo$
1 ^ɯtaʁ/*ɯtaʁ$
1 ^ɯjme/*ɯjme$
1 ^ɕɤrɯ/*ɕɤrɯ$
1 ^xuexiao/*xuexiao$
1 ^tʂu/*tʂu$
1 ^pɯ́wɣsɯmtoa/*pɯ́wɣsɯmtoa$
1 ^mɯpɯrɤʑia/*mɯpɯrɤʑia$
1 ^mɯpjɤpenɯ/*mɯpjɤpenɯ$
1 ^mɤpe/*mɤpe$
1 ^kɯkɯra/*kɯkɯra$
1 ^kɤzo/*kɤzo$
1 ^khramba/*khramba$
1 ^aʑo/*aʑo$
1 ^ambro/*ambro$
coverage: 24 / 42 (~0.57142857142857142857)
remaining unknown forms: 18
ambiguity: 54 / 42 (~1.28571428571428571429)
(JYA ‣ ENG) Transfer Evaluation
We implemented Contrastive Grammar #1: "For clauses with transitive verbs, in Japhug the object generally precedes the verb (and follows the subject). In English however the object follows the verb."
Example 1
lɯlu nɯ kɯ βʑɯ ɯɕa tɕi pjɤrga outputs #that #cat also #like #mouse #meat instead of the direct translation #cat that with #mouse #meat also #like.
Tagger Output
Lex Output
Transfer Output
JYA-ENG Output
#that #cat also #like #mouse #meat
Example 2
βzaŋsa tonɯpandʑi outputs #agree with #friend instead of #friend #agree with.
Tagger Output
^βzaŋsa<n><ind>$ ^nɯpa<v><tv>$^.<sent>$
Lex Output
^βzaŋsa<n><ind>/friend<n><sg><ind>$ ^nɯpa<v><tv>/agree with<vblex>$^.<sent>/.<sent>$
Transfer Output
Applying rule 10 (line 77): ^βzaŋsa<n><ind>/friend<n><sg><ind>$
Applying rule 12 (line 80): ^default<NP>{^βzaŋsa<n><ind>/friend<n><sg><ind>$}$
Applying rule 14 (line 83): ^default<dP><CD>{^default<NP>{^βzaŋsa<n><ind>/friend<n><sg><ind>$}$}$
Applying rule 16 (line 89): ^nɯpa<v><tv>/agree with<vblex>$
Applying rule 17 (line 91): ^default<DP><CD>{^default<dP><CD>{^default<NP>{^βzaŋsa<n><ind>/friend<n><sg><ind>$}$}$}$ ^default<vP>{^nɯpa<v><tv>/agree with<vblex>$}$
Applying output rule 16 (line 91): default<VP> -> ^default<DP><CD>{^default<dP><CD>{^default<NP>{^βzaŋsa<n><ind>/friend<n><sg><ind>$}$}$}$ ^default<vP>{^nɯpa<v><tv>/agree with<vblex>$}$
Applying output rule 15 (line 89): default<vP> -> ^nɯpa<v><tv>/agree with<vblex>$
No rule specified: ^agree with<vblex>$
Applying output rule 13 (line 83): default<DP><abs> -> ^default<dP><CD>{^default<NP>{^βzaŋsa<n><ind>/friend<n><sg><ind>$}$}$
Applying output rule 11 (line 80): default<dP><abs> -> ^default<NP>{^βzaŋsa<n><ind>/friend<n><sg><ind>$}$
Applying output rule 9 (line 77): default<NP> -> ^βzaŋsa<n><ind>/friend<n><sg><ind>$
No rule specified: ^friend<n><sg><ind>$
No rule specified: ^.<sent>/.<sent>$
^agree with<vblex>$ ^friend<n><sg><ind>$^.<sent>$
JYA-ENG Output
#agree with #friend
Our transfer rule successfully switched the object (friend) with the verb phrase (agree with).
Post-Evaluation
Statistics About Input Files
Results when removing unknown-word marks (stars)
Coverage Test on jya.tests.txt in Bilingual Transducer
