Difference between revisions of "Guarani and Warlpiri/Structural transfer"
From LING073
(→Post-Evaluation) |
(→gug → wbp) |
||
| (5 intermediate revisions by 2 users not shown) | |||
| Line 7: | Line 7: | ||
===wbp → gug=== | ===wbp → gug=== | ||
| − | |||
| − | |||
| − | |||
| − | |||
Results when running on gug.sentences.txt: | Results when running on gug.sentences.txt: | ||
*Word error rate (WER): 100.0 % | *Word error rate (WER): 100.0 % | ||
*Position-independent word error rate (PER): 100.0 % | *Position-independent word error rate (PER): 100.0 % | ||
| − | |||
| − | |||
==Structural Transfer Rule== | ==Structural Transfer Rule== | ||
| Line 28: | Line 22: | ||
**apertium-transfer: Rule 1 .<sent>/.<sent> ^n<SN><sg><subj>{^kurdu<n><2>$}$ ^vaux<Saux><SubjTBD>{^ka<vaux><pres><2>$}$ ^v<SV>{^ngarni<v><tv><pres>$}$^sent<SENT>{^.<sent>$}$ | **apertium-transfer: Rule 1 .<sent>/.<sent> ^n<SN><sg><subj>{^kurdu<n><2>$}$ ^vaux<Saux><SubjTBD>{^ka<vaux><pres><2>$}$ ^v<SV>{^ngarni<v><tv><pres>$}$^sent<SENT>{^.<sent>$}$ | ||
*postchunk: ^kurdu<n><sg>$ ^ka<vaux><pres><subj3sg>$ ^ngarni<v><tv><pres>$^.<sent>$ | *postchunk: ^kurdu<n><sg>$ ^ka<vaux><pres><subj3sg>$ ^ngarni<v><tv><pres>$^.<sent>$ | ||
| − | *gug-wbp: #kurdu | + | *gug-wbp: #kurdu ka ngarni |
===wbp → gug=== | ===wbp → gug=== | ||
| − | * | + | *Sample sentence: kurdu ka ngarni → mitã okaru |
| − | * | + | *tagger: ^kurdu<n><sg><abs>$ ^ka<vaux><pres><subj3sg>$ ^ngarni<v><tv><pres>$^.<sent>$ |
| − | + | *biltrans: ^kurdu<n><sg><abs>/mitã<n><sg><abs>$ ^ka<vaux><pres><subj3sg>/<pres><subj3sg>$ ^ngarni<v><tv><pres>/karu<v><tv><ar><pres>$^.<sent>/.<sent>$ | |
| + | *chunker: | ||
| + | **apertium-transfer: Rule 2 ka<vaux><pres><subj3sg>/<pres><subj3sg> ngarni<v><tv><pres>/karu<v><tv><ar><pres> | ||
| + | **apertium-transfer: Rule 1 .<sent>/.<sent> | ||
| + | *interchunk: ^default<default>{^mitã<n><sg><abs>$}$ ^sent<v>{^karu<v><tv><ar><pres><p3><sp>$}$^sent<SENT>{^.<sent>$}$ | ||
| + | *postchunk: ^mitã<n><sg><abs>$ ^karu<v><tv><ar><pres><p3><sp>$^.<sent>$ | ||
| + | *wbp-gug: #mitã okaru (need a transfer rule for nouns to add the subject tag and remove abs tag) | ||
==Post-Evaluation== | ==Post-Evaluation== | ||
| Line 42: | Line 42: | ||
*Position-independent word error rate (PER): 115.52 % | *Position-independent word error rate (PER): 115.52 % | ||
| − | ===gug | + | Results when running on gug.tests.txt: |
| + | *Word error rate (WER): 71.43 % | ||
| + | *Position-independent word error rate (PER): 71.43 % | ||
| + | |||
| + | ===wbp → gug === | ||
Results when running on gug.sentences.txt: | Results when running on gug.sentences.txt: | ||
*Word error rate (WER): 100.00 % | *Word error rate (WER): 100.00 % | ||
*Position-independent word error rate (PER): 100.00 % | *Position-independent word error rate (PER): 100.00 % | ||
| + | (I think nouns are not translating correctly anymore after adding the <subj> tags to them. Will try to fix by next week) | ||
| + | |||
[[Category:Sp17_StructuralTransfer]] | [[Category:Sp17_StructuralTransfer]] | ||
Latest revision as of 16:36, 11 April 2017
Contents
Pre-Evaluation
gug → wbp
Results when running on gug.sentences.txt:
- Word error rate (WER): 113.79 %
- Position-independent word error rate (PER): 113.79 %
wbp → gug
Results when running on gug.sentences.txt:
- Word error rate (WER): 100.0 %
- Position-independent word error rate (PER): 100.0 %
Structural Transfer Rule
gug → wbp
- Sample sentence: mitã okaru → kurdu ka ngarni
- tagger: ^mitã<n><sg><subj>$ ^karu<v><tv><ar><pres><p3><sp>$^.<sent>$
- biltrans: ^mitã<n><sg><subj>/kurdu<n><sg><subj>$ ^karu<v><tv><ar><pres><p3><sp>/ngarni<v><tv><ar><pres><p3><sp>$^.<sent>/.<sent>$
- chunker:
- apertium-transfer: Rule 2 mitã<n><sg><subj>/kurdu<n><sg><subj>
- apertium-transfer: Rule 3 karu<v><tv><ar><pres><p3><sp>/ngarni<v><tv><ar><pres><p3><sp>
- apertium-transfer: Rule 1 .<sent>/.<sent> ^n<SN><sg><subj>{^kurdu<n><2>$}$ ^vaux<Saux><SubjTBD>{^ka<vaux><pres><2>$}$ ^v<SV>{^ngarni<v><tv><pres>$}$^sent<SENT>{^.<sent>$}$
- postchunk: ^kurdu<n><sg>$ ^ka<vaux><pres><subj3sg>$ ^ngarni<v><tv><pres>$^.<sent>$
- gug-wbp: #kurdu ka ngarni
wbp → gug
- Sample sentence: kurdu ka ngarni → mitã okaru
- tagger: ^kurdu<n><sg><abs>$ ^ka<vaux><pres><subj3sg>$ ^ngarni<v><tv><pres>$^.<sent>$
- biltrans: ^kurdu<n><sg><abs>/mitã<n><sg><abs>$ ^ka<vaux><pres><subj3sg>/<pres><subj3sg>$ ^ngarni<v><tv><pres>/karu<v><tv><ar><pres>$^.<sent>/.<sent>$
- chunker:
- apertium-transfer: Rule 2 ka<vaux><pres><subj3sg>/<pres><subj3sg> ngarni<v><tv><pres>/karu<v><tv><ar><pres>
- apertium-transfer: Rule 1 .<sent>/.<sent>
- interchunk: ^default<default>{^mitã<n><sg><abs>$}$ ^sent<v>{^karu<v><tv><ar><pres><p3><sp>$}$^sent<SENT>{^.<sent>$}$
- postchunk: ^mitã<n><sg><abs>$ ^karu<v><tv><ar><pres><p3><sp>$^.<sent>$
- wbp-gug: #mitã okaru (need a transfer rule for nouns to add the subject tag and remove abs tag)
Post-Evaluation
gug → wbp
Results when running on gug.sentences.txt:
- Word error rate (WER): 115.52 %
- Position-independent word error rate (PER): 115.52 %
Results when running on gug.tests.txt:
- Word error rate (WER): 71.43 %
- Position-independent word error rate (PER): 71.43 %
wbp → gug
Results when running on gug.sentences.txt:
- Word error rate (WER): 100.00 %
- Position-independent word error rate (PER): 100.00 %
(I think nouns are not translating correctly anymore after adding the <subj> tags to them. Will try to fix by next week)
