Difference between revisions of "Magahi and English/Structural transfer"

From LING073
Jump to: navigation, search
(Pre-Evaluation)
 
(23 intermediate revisions by the same user not shown)
Line 2: Line 2:
 
Statistics about input files
 
Statistics about input files
 
-------------------------------------------------------
 
-------------------------------------------------------
Number of words in reference: 52
+
Number of words in reference: 52 <br>
Number of words in test: 42
+
Number of words in test: 42 <br>
Number of unknown words (marked with a star) in test: 10
+
Number of unknown words (marked with a star) in test: 10 <br>
Percentage of unknown words: 23.81 %
+
Percentage of unknown words: 23.81 % <br>
  
 
Results when removing unknown-word marks (stars)
 
Results when removing unknown-word marks (stars)
 
-------------------------------------------------------
 
-------------------------------------------------------
Edit distance: 52
+
Edit distance: 52 <br>
Word error rate (WER): 100.00 %
+
Word error rate (WER): 100.00 % <br>
Number of position-independent correct words: 2
+
Number of position-independent correct words: 2 <br>
Position-independent word error rate (PER): 96.15 %
+
Position-independent word error rate (PER): 96.15 % <br>
 +
 
 +
= Examples =
 +
 
 +
== Postpositions ==
 +
 
 +
Tagger:
 +
^pāoṃ<n>$ ^lāg<v><pres>$ ^ke<post>$^.<sent>$^.<sent>
 +
 
 +
Lex:
 +
^pāoṃ<n>/feet<n>$ ^lāg<v><pres>/touch<vblex><pres>$ ^ke<post>/that<cnjsub>$^.<sent>/.<sent>$^.<sent>/.<sent>$
 +
 
 +
Transfer: 
 +
Applying rule 5 (line 59): ^pāoṃ<n>/feet<n>$
 +
Applying rule 3 (line 57): ^default<NP><ND>{^pāoṃ<n>/feet<n>$}$
 +
Applying rule 9 (line 64): ^lāg<v><pres>/touch<vblex><pres>$
 +
Applying rule 8 (line 62): ^default<vP><SD>{^lāg<v><pres>/touch<vblex><pres>$}$
 +
Applying rule 1 (line 55): ^default<DP><PD><ND>{^default<NP><ND>{^pāoṃ<n>/feet<n>$}$}$ ^default<VP><SD>{^default<vP><SD>{^lāg<v><pres>/touch<vblex><pres>$}$}$
 +
Applying output rule 0 (line 55): default{{tag|S|o}} -> ^default<DP><PD><ND>{^default<NP><ND>{^pāoṃ<n>/feet<n>$}$}$ ^default<VP><SD>{^default<vP><SD>{^lāg<v><pres>/touch<vblex><pres>$}$}$
 +
Applying output rule 2 (line 57): default<DP><p3><sg> -> ^default<NP><ND>{^pāoṃ<n>/feet<n>$}$
 +
No rule specified: ^the<det><def><sp>$
 +
Applying output rule 4 (line 59): default<NP><sg> -> ^pāoṃ<n>/feet<n>$
 +
No rule specified: ^feet<n><sg>$
 +
Applying output rule 7 (line 62): default<VP><s_p1> -> ^default<vP><SD>{^lāg<v><pres>/touch<vblex><pres>$}$
 +
Applying output rule 8 (line 64): default<vP><s_p1> -> ^lāg<v><pres>/touch<vblex><pres>$
 +
No rule specified: ^touch<vblex><pres>$
 +
No rule specified: ^ke<post>/that<cnjsub>$
 +
No rule specified: ^.<sent>/.<sent>$
 +
No rule specified: ^.<sent>/.<sent>$^the<det><def><sp>$ ^feet<n><sg>$ ^touch<vblex><pres>$ ^that<cnjsub>$^.<sent>$^.<sent>
 +
 
 +
mag-eng:
 +
the #feet touch that
 +
 
 +
== Verb Conjugation ==
 +
{{transferTest|mag|eng|Ham sutli.|I slept.}} <br>
 +
Tagger:
 +
^Prpers<prn><pers><p1><sg>$ ^*sutli$^.<sent>$
 +
Lex:
 +
^Prpers<prn><pers><p1><sg>/Prpers<prn><pers><p1><sg>$ ^sut<v><past><s_p1>/sleep<vblex><past><s_p1>$^.<sent>/.<sent>$^.<sent>/.<sent>
 +
 
 +
Transfer: 
 +
No rule specified: ^Prpers<prn><pers><p1><sg>/Prpers<prn><pers><p1><sg>$
 +
Applying rule 9 (line 64): ^sut<v><past><s_p1>/sleep<vblex><past><s_p1>$
 +
Applying rule 8 (line 62): ^default<vP><s_p1>{^sut<v><past><s_p1>/sleep<vblex><past><s_p1>$}$
 +
Applying output rule 7 (line 62): default<VP><s_p1> -> ^default<vP><s_p1>{^sut<v><past><s_p1>/sleep<vblex><past><s_p1>$}$
 +
Applying output rule 8 (line 64): default<vP><s_p1> -> ^sut<v><past><s_p1>/sleep<vblex><past><s_p1>$
 +
No rule specified: ^sleep<vblex><past>$
 +
No rule specified: ^.<sent>/.<sent>$
 +
No rule specified: ^.<sent>/.<sent>$
 +
^Prpers<prn><pers><p1><sg>$ ^sleep<vblex><past>$^.<sent>$^.<sent>
 +
 
 +
mag-eng:
 +
#Prpers slept
 +
 
 +
== Verb Tenses ==
 +
{{transferTest|mag|eng|Sādhū dekhbo.|The saint will see.}}
 +
Tagger:
 +
^Sādhū<n>$ ^dekh<v><fut><s_p3>$^.<sent>$
 +
 
 +
Lex:
 +
^Sādhū<n>/Saint<n>$ ^dekh<v><fut><s_p3>/see<vblex><fut><s_p3>$^.<sent>/.<sent>$^.<sent>/.<sent>
 +
 
 +
Transfer: 
 +
Applying rule 5 (line 59): ^Sādhū<n>/Saint<n>$
 +
Applying rule 3 (line 57): ^Default<NP><ND>{^Sādhū<n>/Saint<n>$}$
 +
Applying rule 9 (line 64): ^dekh<v><fut><s_p3>/see<vblex><fut><s_p3>$
 +
Applying rule 8 (line 62): ^default<vP><s_p3>{^dekh<v><fut><s_p3>/see<vblex><fut><s_p3>$}$
 +
Applying rule 1 (line 55): ^Default<DP><PD><ND>{^Default<NP><ND>{^Sādhū<n>/Saint<n>$}$}$ ^default<VP><s_p3>{^default<vP><s_p3>{^dekh<v><fut><s_p3>/see<vblex><fut><s_p3>$}$}$
 +
Applying output rule 0 (line 55): Default{{tag|S|o}} -> ^Default<DP><PD><ND>{^Default<NP><ND>{^Sādhū<n>/Saint<n>$}$}$ ^default<VP><s_p3>{^default<vP><s_p3>{^dekh<v><fut><s_p3>/see<vblex><fut><s_p3>$}$}$
 +
Applying output rule 2 (line 57): Default<DP><p3><sg> -> ^Default<NP><ND>{^Sādhū<n>/Saint<n>$}$
 +
No rule specified: ^the<det><def><sp>$
 +
Applying output rule 4 (line 59): Default<NP><sg> -> ^Sādhū<n>/Saint<n>$
 +
No rule specified: ^Saint<n><sg>$
 +
Applying output rule 7 (line 62): default<VP><s_p3> -> ^default<vP><s_p3>{^dekh<v><fut><s_p3>/see<vblex><fut><s_p3>$}$
 +
Applying output rule 8 (line 64): default<vP><s_p3> -> ^dekh<v><fut><s_p3>/see<vblex><fut><s_p3>$
 +
No rule specified: ^will<vbmod><pres>$
 +
No rule specified: ^see<vblex><inf>$
 +
No rule specified: ^.<sent>/.<sent>$
 +
No rule specified: ^.<sent>/.<sent>$
 +
^the<det><def><sp>$ ^Saint<n><sg>$ ^will<vbmod><pres>$ ^see<vblex><inf>$^.<sent>$^.<sent>
 +
 
 +
mag-eng:
 +
the Saint will see.
 +
 
 +
= Post-Evaluation =
 +
Number of words in reference: 69 <br>
 +
Number of words in test: 75 <br>
 +
Number of unknown words (marked with a star) in test: 8 <br>
 +
Percentage of unknown words: 10.67 % <br>
 +
 
 +
Results when removing unknown-word marks (stars)
 +
-------------------------------------------------------
 +
Edit distance: 57 <br>
 +
Word error rate (WER): 82.61 % <br>
 +
Number of position-independent correct words: 22 <br>
 +
Position-independent word error rate (PER): 76.81 % <br>
 +
 
 +
Results when unknown-word marks (stars) are not removed
 +
-------------------------------------------------------
 +
Edit distance: 57 <br>
 +
Word Error Rate (WER): 82.61 % <br>
 +
Number of position-independent correct words: 22 <br>
 +
Position-independent word error rate (PER): 76.81 % <br>
  
 
[[Category:Magahi]][[Category:English]][[Category:Sp21_StructuralTransfer]]
 
[[Category:Magahi]][[Category:English]][[Category:Sp21_StructuralTransfer]]

Latest revision as of 22:45, 3 May 2021

Pre-Evaluation

Statistics about input files


Number of words in reference: 52
Number of words in test: 42
Number of unknown words (marked with a star) in test: 10
Percentage of unknown words: 23.81 %

Results when removing unknown-word marks (stars)


Edit distance: 52
Word error rate (WER): 100.00 %
Number of position-independent correct words: 2
Position-independent word error rate (PER): 96.15 %

Examples

Postpositions

Tagger:

^pāoṃ<n>$ ^lāg<v><pres>$ ^ke<post>$^.<sent>$^.<sent>

Lex:

^pāoṃ<n>/feet<n>$ ^lāg<v><pres>/touch<vblex><pres>$ ^ke<post>/that<cnjsub>$^.<sent>/.<sent>$^.<sent>/.<sent>$

Transfer:

Applying rule 5 (line 59): ^pāoṃ<n>/feet<n>$
Applying rule 3 (line 57): ^default<NP><ND>{^pāoṃ<n>/feet<n>$}$
Applying rule 9 (line 64): ^lāg<v><pres>/touch<vblex><pres>$
Applying rule 8 (line 62): ^default<vP><SD>{^lāg<v><pres>/touch<vblex><pres>$}$
Applying rule 1 (line 55): ^default<DP><PD><ND>{^default<NP><ND>{^pāoṃ<n>/feet<n>$}$}$ ^default<VP><SD>{^default<vP><SD>{^lāg<v><pres>/touch<vblex><pres>$}$}$
Applying output rule 0 (line 55): default<S> -> ^default<DP><PD><ND>{^default<NP><ND>{^pāoṃ<n>/feet<n>$}$}$ ^default<VP><SD>{^default<vP><SD>{^lāg<v><pres>/touch<vblex><pres>$}$}$
Applying output rule 2 (line 57): default<DP><p3><sg> -> ^default<NP><ND>{^pāoṃ<n>/feet<n>$}$
No rule specified: ^the<det><def><sp>$
Applying output rule 4 (line 59): default<NP><sg> -> ^pāoṃ<n>/feet<n>$
No rule specified: ^feet<n><sg>$
Applying output rule 7 (line 62): default<VP><s_p1> -> ^default<vP><SD>{^lāg<v><pres>/touch<vblex><pres>$}$
Applying output rule 8 (line 64): default<vP><s_p1> -> ^lāg<v><pres>/touch<vblex><pres>$
No rule specified: ^touch<vblex><pres>$
No rule specified: ^ke<post>/that<cnjsub>$
No rule specified: ^.<sent>/.<sent>$
No rule specified: ^.<sent>/.<sent>$^the<det><def><sp>$ ^feet<n><sg>$ ^touch<vblex><pres>$ ^that<cnjsub>$^.<sent>$^.<sent>

mag-eng: the #feet touch that

Verb Conjugation

(mag) Ham sutli. → (eng) I slept.
Tagger:

^Prpers<prn><pers><p1><sg>$ ^*sutli$^.<sent>$

Lex:

^Prpers<prn><pers><p1><sg>/Prpers<prn><pers><p1><sg>$ ^sut<v><past><s_p1>/sleep<vblex><past><s_p1>$^.<sent>/.<sent>$^.<sent>/.<sent>

Transfer:

No rule specified: ^Prpers<prn><pers><p1><sg>/Prpers<prn><pers><p1><sg>$
Applying rule 9 (line 64): ^sut<v><past><s_p1>/sleep<vblex><past><s_p1>$
Applying rule 8 (line 62): ^default<vP><s_p1>{^sut<v><past><s_p1>/sleep<vblex><past><s_p1>$}$
Applying output rule 7 (line 62): default<VP><s_p1> -> ^default<vP><s_p1>{^sut<v><past><s_p1>/sleep<vblex><past><s_p1>$}$
Applying output rule 8 (line 64): default<vP><s_p1> -> ^sut<v><past><s_p1>/sleep<vblex><past><s_p1>$
No rule specified: ^sleep<vblex><past>$
No rule specified: ^.<sent>/.<sent>$
No rule specified: ^.<sent>/.<sent>$
^Prpers<prn><pers><p1><sg>$ ^sleep<vblex><past>$^.<sent>$^.<sent>

mag-eng:

#Prpers slept

Verb Tenses

(mag) Sādhū dekhbo. → (eng) The saint will see. Tagger:

^Sādhū<n>$ ^dekh<v><fut><s_p3>$^.<sent>$

Lex:

^Sādhū<n>/Saint<n>$ ^dekh<v><fut><s_p3>/see<vblex><fut><s_p3>$^.<sent>/.<sent>$^.<sent>/.<sent>

Transfer:

Applying rule 5 (line 59): ^Sādhū<n>/Saint<n>$
Applying rule 3 (line 57): ^Default<NP><ND>{^Sādhū<n>/Saint<n>$}$
Applying rule 9 (line 64): ^dekh<v><fut><s_p3>/see<vblex><fut><s_p3>$
Applying rule 8 (line 62): ^default<vP><s_p3>{^dekh<v><fut><s_p3>/see<vblex><fut><s_p3>$}$
Applying rule 1 (line 55): ^Default<DP><PD><ND>{^Default<NP><ND>{^Sādhū<n>/Saint<n>$}$}$ ^default<VP><s_p3>{^default<vP><s_p3>{^dekh<v><fut><s_p3>/see<vblex><fut><s_p3>$}$}$
Applying output rule 0 (line 55): Default<S> -> ^Default<DP><PD><ND>{^Default<NP><ND>{^Sādhū<n>/Saint<n>$}$}$ ^default<VP><s_p3>{^default<vP><s_p3>{^dekh<v><fut><s_p3>/see<vblex><fut><s_p3>$}$}$
Applying output rule 2 (line 57): Default<DP><p3><sg> -> ^Default<NP><ND>{^Sādhū<n>/Saint<n>$}$
No rule specified: ^the<det><def><sp>$
Applying output rule 4 (line 59): Default<NP><sg> -> ^Sādhū<n>/Saint<n>$
No rule specified: ^Saint<n><sg>$
Applying output rule 7 (line 62): default<VP><s_p3> -> ^default<vP><s_p3>{^dekh<v><fut><s_p3>/see<vblex><fut><s_p3>$}$
Applying output rule 8 (line 64): default<vP><s_p3> -> ^dekh<v><fut><s_p3>/see<vblex><fut><s_p3>$
No rule specified: ^will<vbmod><pres>$
No rule specified: ^see<vblex><inf>$
No rule specified: ^.<sent>/.<sent>$
No rule specified: ^.<sent>/.<sent>$
^the<det><def><sp>$ ^Saint<n><sg>$ ^will<vbmod><pres>$ ^see<vblex><inf>$^.<sent>$^.<sent>

mag-eng:

the Saint will see.

Post-Evaluation

Number of words in reference: 69
Number of words in test: 75
Number of unknown words (marked with a star) in test: 8
Percentage of unknown words: 10.67 %

Results when removing unknown-word marks (stars)


Edit distance: 57
Word error rate (WER): 82.61 %
Number of position-independent correct words: 22
Position-independent word error rate (PER): 76.81 %

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


Edit distance: 57
Word Error Rate (WER): 82.61 %
Number of position-independent correct words: 22
Position-independent word error rate (PER): 76.81 %