Evenki and English/Structural transfer - Revision history

Jwashin1 at 18:34, 11 June 2023

2023-06-11T18:34:46Z

2023-05-11T16:54:16Z

‎Mode Outputs For Each Sentence

2023-05-11T16:54:05Z

‎Structural Transfer Rules

2023-05-11T14:49:33Z

2023-05-11T14:48:09Z

2023-05-11T13:24:04Z

‎Pre-Evaluation

2023-05-11T13:09:10Z

‎Pre-Evaluation

2023-05-11T13:05:40Z

‎Pre-Evaluation

2023-05-10T00:11:13Z

2023-05-10T00:10:38Z

← Older revision		Revision as of 18:34, 11 June 2023
Line 47:		Line 47:

	Note: Nothing improved when implementing structural transfer but this is likely due to the same issue from the transducer preventing words in both monolingual and bilingual dictionaries from mapping to any translation. Should that error be resolved significant improvement would likely be seen however fixing that issue is currently outside of the scope of our capabilities.		Note: Nothing improved when implementing structural transfer but this is likely due to the same issue from the transducer preventing words in both monolingual and bilingual dictionaries from mapping to any translation. Should that error be resolved significant improvement would likely be seen however fixing that issue is currently outside of the scope of our capabilities.
		+
		+	{{comment\|No, the rules were just put in the wrong file}}

	[[Category:Sp23_StructuralTransfer ]][[Category:evenki]][[Category:english]]		[[Category:Sp23_StructuralTransfer ]][[Category:evenki]][[Category:english]]

@@ Line 22: / Line 22: @@
 === Mode Outputs For Each Sentence===
 {| class="wikitable"
 |-

@@ Line 19: / Line 19: @@
 === Structural Transfer Rules ===
-     We added a structural transfer rule to account for the different subject complement verb orders in Evenki and English.
+We added a structural transfer rule to account for the different subject complement verb orders in Evenki and English.
 === Mode Outputs For Each Sentence===

@@ Line 49: / Line 49: @@
   Position-independent word error rate (PER): 110.00 %
-  Note: Nothing improved when implementing structural transfer but this is likely due to the same orthographical issue from the transducer. Should
+  Note: Nothing improved when implementing structural transfer but this is likely due to the same issue from the transducer preventing words in both monolingual and bilingual dictionaries from mapping to any translation. Should that error be resolved significant improvement would likely be seen however fixing that issue is currently outside of the scope of our capabilities.
 [[Category:Sp23_StructuralTransfer ]][[Category:evenki]][[Category:english]]

@@ Line 15: / Line 15: @@
   Number of position-independent correct words: 0
   Position-independent word error rate (PER): 110.00 %
 === Structural Transfer Rules ===
-      Implement one item from your contrastive grammar.
+      We added a structural transfer rule to account for the different subject complement verb orders in Evenki and English.
 === Mode Outputs For Each Sentence===
@@ Line 38: / Line 29: @@
 ! sentence !! tagger !! lex !! transfer !! evn-eng
 |-
-| sentence || Output || Output || Output || Output
+| Тадук нунгартын аичимни чатын. || ^*Тадук$ ^*нунгартын$ ^*аичимни$ ^*чатын$. || ^*Тадук/*Тадук$ ^*нунгартын/*нунгартын$ ^*аичимни/*аичимни$ ^*чатын/*чатын$. || ^*Тадук$ ^*нунгартын$ ^*аичимни$ ^*чатын$. || *Тадук *нунгартын *аичимни *чатын.
 |-
 |}
@@ Line 49: / Line 36: @@
-{| class="wikitable"
+ -------------------------------------------------------
-|-
+ Number of words in reference: 20
-! WER Test !! PER Test !! Coverage Test
+ Number of words in test: 22
-|-
+ Number of unknown words (marked with a star) in test: 20
-| Output || Output || Output
+ Percentage of unknown words: 90.91 %
-|-
-|}
+ Results when removing unknown-word marks (stars)
 [[Category:Sp23_StructuralTransfer ]][[Category:evenki]][[Category:english]]

@@ Line 1: / Line 1: @@
 === Pre-Evaluation ===
-  Note: Due to the previously mentioned orthography issues with the Evenki transducer, our test results are artificially low. Since the transducer is imported from apertium and we did not created, this currently not a problem we can fix. However, our work could be applied in future work on evenki if this issue were to be fixed.
+  Note: Due to the previously mentioned orthography issues with the Evenki transducer, our test results are artificially low. Since the transducer is imported from apertium and we did not created, this currently not a problem we can fix. However, our work could be applied in future work on Evenki if this issue were to be fixed.
 {| class="wikitable"

← Older revision		Revision as of 00:11, 10 May 2023
Line 1:		Line 1:
−	=== Pre=Evaluation ===	+	=== Pre-Evaluation ===

	Before you start development of structural transfer, perform WER, PER, and coverage tests on your short sentences corpus, and add this in to a pre-evaluation section.		Before you start development of structural transfer, perform WER, PER, and coverage tests on your short sentences corpus, and add this in to a pre-evaluation section.