LING073 - User contributions [en]

Hwang11/Final project

2018-05-17T18:01:22Z

Hwang11:

=Overview=
This final project is an extension of the last assignment on universal dependencies. I continue working on the language, FIjian, I chose at the beginning. For this project, I plan to annotate 232 sentences in a text drawn from the main grammar book ''A Grammar of Boumaa Fijian'' by R.M.W. Dixon (currently only 110 sentences are annotated). In this process, I also expect to enrich my morphological transduce as encountering more sentences and new words. Then after annotating the sentences, I will also train a parser and evaluate the accuracy of the parser on both the data used for training and other data. Through annotating a larger corpus, I hope to develop a more accurate parser than the one for my last assignment and also to explore more about the dependency syntax in Fijian.

=Evaluation=
The current evaluation is on the transducer with 260 more stems added and a model trained by a <code>.conllu</code> file containing 90 annotated sentences from the text in the grammar book. The "other data" used in the test right now is the fij.annotated.ud.conllu annotated in the previous assignment.
*Current coverage over a large corpus: 75.26%
:-This rate is expected because though a lot more stems are added into the transducer, they are almost all in Boumaa Fijian, a dialect overlapping 80% with the Standard Fijian, which the large corpus (Bible New Testament) is in.
* Accuracy for the trained UD-relations parser:
{| class="wikitable"
!
! fij.withmorph.udpipe
! fij.nomorph.udpipe
! Number of forms
! Number of sentences
|-
! Training data
| UAS: 96.83%, LAS: 94.96%
| UAS: 86.02%, LAS: 82.65%
| 1481
| 90
|-
! Other data
| UAS: 73.05%, LAS: 59.53%
| UAS: 67.02%, LAS: 57.80%
| 282
| 30
|}
The relatively low accuracy for the models on the other data is expected, since that corpus also includes several sentences in Standard Fijian.
=UD Guidelines=
This section lists several frequently used dependencies in Fijian, including a description and one or two examples for each of them.
==root==
In Fijian, not only verb, but any Part of Speech except time words can be a predicate head, and thus can possibly be a root.
*Example: number as the root
[[File:root.png|200px|thumb|left|'There was once a king in this land of Boumaa, the Vuunisaa.']]

==nsubj, expl, csubj==
A predicate usually requires a subject pronoun preceding the predicate head (but it can be omitted sometimes, e.g. when a cardinal pronoun is present following the predicate head), and subject NP and subject clause (with the same reference as the subject pronoun) always follow the predicate. When a subject NP or a subject clause is present in a clause, the subject pronoun gets<expl>, and the subject NP gets <nsubj> or the subject clause gets <csubj>. When neither is present but only the subject pronoun is present in the clause, the subject pronoun gets <nsubj> instead of <expl>.

For <expl>, the dependent is a pronoun (subj, obj, card); for <nsubj>, the dependent must be a nominal; for <csubj>, the dependent must be a predicate head.
*Example: A clause containing all three dependencies. For the matrix clause, the subject is a clause and within the subject clause, the subject is a noun.
[[File:expl-nsubj-csubj.png|200px|thumb|left|'One of the women said:']]

==obl, advmod, advcl==
These three are all modifiers of a predicate.

Oblique <obl> is usually a non-core nominal adjunct of the predicate, and in Fijian, it is usually preceded by a preposition. The dependent must be a noun.

A predicate usually takes several adverbial modifiers in a clause, and the <advmod> does not have to be an adverb (e.g. deictic verbs can modify a predicate as well.)
*Example:
[[File:obl-advmod.png|200px|thumb|left|'The two of them went counter-clockwise, reached Vurevure.']]

Adverbial clause modifier is a clause that modifies a predicate. The dependent is the predicate head of the adv clause. The clause is usually temporal clause, causal clause, conditional clause, etc..
*Example: causal clause
[[File:advcl.png|200px|thumb|left|'Well, he was known because of the unique length of his feet (lit: because his feet is very long).']]

==aux==
Fijian has separate morphemes as aspect and tense markers, which all get <aux> relation with the predicate head.
*Example: When an <aux> is present, the subject pronoun is not required.
[[File:aux.png|200px|thumb|left|'The married couple wanted to sleep.']]

==xcomp==
An open clause with its own subject. The reference of the subject of <xcomp> is usually the subject or object of the matrix clause. The dependent must be on a predicate head.
[[File:aux.png|200px|thumb|left|'The married couple wanted to sleep.']]

==obj, ccomp, iobj==
Object and ccomp are not very common in Fijian, since 70% Fijian clauses are intransitive. No indirect object has been found in my corpus yet.

=Issues and Future Work=
* The current annotations are mainly on Boumaa Fijian. Although it is very similar to Standard Fijian in syntax, it is still necessary and worthwhile to annotate more sentences in Standard Fijian.

* The grammar book I am using does not distinguish adverbs and aspect markers very well and put them all into the "modifier" category. Some words are said to be aspect markers, but their corresponding translations in English are like adverbs, such as 'continuously'. When developing my transducers, I added tags according their translations (i.e. I gave adv tag to some words claimed to be asp markers by the grammar book) but during the process of doing UD relations I realised that for future work, it might be necessary and important to better distinguish asp markers and true adverbs.

=License and Copyright=
The annotation is licensed under an open source license GNU GLP v3. The original text is protected by copyright: All rights reserved by R.M.W. Dixion, the author of the book ''A grammar of Boumaa Fijian''.

=References and Resources=
For more details: https://github.swarthmore.edu/hwang11/fij-final/tree/master.

And it is also published publicly on github: https://github.com/hwang16/UD_fij.

Churchward, C. M. (1941). A new Fijian grammar. Suva (Fiji): Government of Fiji.

Dixon, R.M.W. (1988). A Grammar of Boumaa Fijian. Chicago: University of Chicago Press.

[[Category:sp18_FinalProjects]]

Hwang11/Final project

2018-05-17T13:37:43Z

Hwang11:

=Overview=
This final project is an extension of the last assignment on universal dependencies. I continue working on the language, FIjian, I chose at the beginning. For this project, I plan to annotate 232 sentences in a text drawn from the main grammar book ''A Grammar of Boumaa Fijian'' by R.M.W. Dixon. In this process, I also expect to enrich my morphological transduce as encountering more sentences and new words. Then after annotating the sentences, I will also train a parser and evaluate the accuracy of the parser on both the data used for training and other data. Through annotating a larger corpus, I hope to develop a more accurate parser than the one for my last assignment and also to explore more about the dependency syntax in Fijian.

=Evaluation=
The current evaluation is on the transducer with 260 more stems added and a model trained by a <code>.conllu</code> file containing 90 annotated sentences from the text in the grammar book. The "other data" used in the test right now is the fij.annotated.ud.conllu annotated in the previous assignment.
*Current coverage over a large corpus: 75.26%
:-This rate is expected because though a lot more stems are added into the transducer, they are almost all in Boumaa Fijian, a dialect overlapping 80% with the Standard Fijian, which the large corpus (Bible New Testament) is in.
* Accuracy for the trained UD-relations parser:
{| class="wikitable"
!
! fij.withmorph.udpipe
! fij.nomorph.udpipe
! Number of forms
! Number of sentences
|-
! Training data
| UAS: 96.83%, LAS: 94.96%
| UAS: 86.02%, LAS: 82.65%
| 1481
| 90
|-
! Other data
| UAS: 73.05%, LAS: 59.53%
| UAS: 67.02%, LAS: 57.80%
| 282
| 30
|}
The relatively low accuracy for the models on the other data is expected, since that corpus also includes several sentences in Standard Fijian.
=UD Guidelines=
This section lists several frequently used dependencies in Fijian, including a description and one or two examples for each of them.
==root==
In Fijian, not only verb, but any Part of Speech except time words can be a predicate head, and thus can possibly be a root.
*Example: number as the root
[[File:root.png|200px|thumb|left|'There was once a king in this land of Boumaa, the Vuunisaa.']]

==nsubj, expl, csubj==
A predicate usually requires a subject pronoun preceding the predicate head (but it can be omitted sometimes, e.g. when a cardinal pronoun is present following the predicate head), and subject NP and subject clause (with the same reference as the subject pronoun) always follow the predicate. When a subject NP or a subject clause is present in a clause, the subject pronoun gets<expl>, and the subject NP gets <nsubj> or the subject clause gets <csubj>. When neither is present but only the subject pronoun is present in the clause, the subject pronoun gets <nsubj> instead of <expl>.

For <expl>, the dependent is a pronoun (subj, obj, card); for <nsubj>, the dependent must be a nominal; for <csubj>, the dependent must be a predicate head.
*Example: A clause containing all three dependencies. For the matrix clause, the subject is a clause and within the subject clause, the subject is a noun.
[[File:expl-nsubj-csubj.png|200px|thumb|left|'One of the women said:']]

==obl, advmod, advcl==
These three are all modifiers of a predicate.

Oblique <obl> is usually a non-core nominal adjunct of the predicate, and in Fijian, it is usually preceded by a preposition. The dependent must be a noun.

A predicate usually takes several adverbial modifiers in a clause, and the <advmod> does not have to be an adverb (e.g. deictic verbs can modify a predicate as well.)
*Example:
[[File:obl-advmod.png|200px|thumb|left|'The two of them went counter-clockwise, reached Vurevure.']]

Adverbial clause modifier is a clause that modifies a predicate. The dependent is the predicate head of the adv clause. The clause is usually temporal clause, causal clause, conditional clause, etc..
*Example: causal clause
[[File:advcl.png|200px|thumb|left|'Well, he was known because of the unique length of his feet (lit: because his feet is very long).']]

==aux==
Fijian has separate morphemes as aspect and tense markers, which all get <aux> relation with the predicate head.
*Example: When an <aux> is present, the subject pronoun is not required.
[[File:aux.png|200px|thumb|left|'The married couple wanted to sleep.']]

==xcomp==
An open clause with its own subject. The reference of the subject of <xcomp> is usually the subject or object of the matrix clause. The dependent must be on a predicate head.
[[File:aux.png|200px|thumb|left|'The married couple wanted to sleep.']]

==obj, ccomp, iobj==
Object and ccomp are not very common in Fijian, since 70% Fijian clauses are intransitive. No indirect object has been found in my corpus yet.

=Issues and Future Work=
* The current annotations are mainly on Boumaa Fijian. Although it is very similar to Standard Fijian in syntax, it is still necessary and worthwhile to annotate more sentences in Standard Fijian.

* The grammar book I am using does not distinguish adverbs and aspect markers very well and put them all into the "modifier" category. Some words are said to be aspect markers, but their corresponding translations in English are like adverbs, such as 'continuously'. When developing my transducers, I added tags according their translations (i.e. I gave adv tag to some words claimed to be asp markers by the grammar book) but during the process of doing UD relations I realised that for future work, it might be necessary and important to better distinguish asp markers and true adverbs.

=License and Copyright=
The annotation is licensed under an open source license GNU GLP v3. The original text is protected by copyright: All rights reserved by R.M.W. Dixion, the author of the book ''A grammar of Boumaa Fijian''.

=References and Resources=
For more details: https://github.swarthmore.edu/hwang11/fij-final/tree/master.

And it is also published publicly on github: https://github.com/hwang16/UD_fij.

Churchward, C. M. (1941). A new Fijian grammar. Suva (Fiji): Government of Fiji.

Dixon, R.M.W. (1988). A Grammar of Boumaa Fijian. Chicago: University of Chicago Press.

[[Category:sp18_FinalProjects]]

Hwang11/Final project

2018-05-14T14:40:11Z

Fijian and English

2018-05-05T17:42:23Z

Hwang11: /* Additions */

Resources for machine translation between [https://wikis.swarthmore.edu/ling073/Fijian Fijian] and English
==fij → eng evaluation==
Current WER and PER:
<pre>
Test file: 'fij-eng.tests.txt'
Reference file 'eng.tests.txt'

Statistics about input files
-------------------------------------------------------
Number of words in reference: 56
Number of words in test: 59
Number of unknown words (marked with a star) in test:
Percentage of unknown words: 0.00 %

Results when removing unknown-word marks (stars)
-------------------------------------------------------
Edit distance: 50
Word error rate (WER): 89.29 %
Number of position-independent correct words: 12
Position-independent word error rate (PER): 83.93 %

Results when unknown-word marks (stars) are not removed
-------------------------------------------------------
Edit distance: 50
Word Error Rate (WER): 89.29 %
Number of position-independent correct words: 12
Position-independent word error rate (PER): 83.93 %

Statistics about the translation of unknown words
-------------------------------------------------------
Number of unknown words which were free rides: 0
Percentage of unknown words that were free rides: 0%

</pre>

==eng → fij evaluation==
Current WER and PER :
<pre>
Test file: 'eng-fij.tests.txt'
Reference file 'fij.tests.txt'

Statistics about input files
-------------------------------------------------------
Number of words in reference: 62
Number of words in test: 56
Number of unknown words (marked with a star) in test: 2
Percentage of unknown words: 3.57 %

Results when removing unknown-word marks (stars)
-------------------------------------------------------
Edit distance: 52
Word error rate (WER): 83.87 %
Number of position-independent correct words: 13
Position-independent word error rate (PER): 79.03 %

Results when unknown-word marks (stars) are not removed
-------------------------------------------------------
Edit distance: 52
Word Error Rate (WER): 83.87 %
Number of position-independent correct words: 13
Position-independent word error rate (PER): 79.03 %

Statistics about the translation of unknown words
-------------------------------------------------------
Number of unknown words which were free rides: 0
Percentage of unknown words that were free rides: 0.00 %

</pre>

==Lexical Selection==
https://wikis.swarthmore.edu/ling073/Fijian_and_English/Lexical_selection#fij_.E2.86.92_eng
===eng → fij one-to-many mapping===
*Case 1: ''Pelu'' and ''lo’i'' describe two different kinds of bending action.

{{transferTest|eng|fij|bend|pelu}} (e.g. bend of metal)

{{transferTest|eng|fij|bend|lo’i}} (e.g. bend at a joint)

*Case 2:

{{transferTest|eng|fij|shine on|cina}} (light/torch shines on)

{{transferTest|eng|fij|shine on|cila}} (sun/moon/star shines on)

*(a disambiguation problem) The third person singular pronoun in English does not distinguish between nominative and accusative case.

{{transferTest|eng|fij|it|e}} (subj)

{{transferTest|eng|fij|it|koya}} (obj)

===fij → eng one-to-many mapping===
*Case 1.1:

{{transferTest|fij|eng|yava|leg}}

{{transferTest|fij|eng|yava|foot}}

*Case 1.2:

{{transferTest|fij|eng|liga|arm}}

{{transferTest|fij|eng|liga|hand}}

*Case 1.3:

{{transferTest|fij|eng|mata|face}}

{{transferTest|fij|eng|mata|eye}}

*Case 2:

{{transferTest|fij|eng|vula|moon}}

{{transferTest|fij|eng|vula|month}}

*Case 3:

{{transferTest|fij|eng|basu|tear up}} (e.g. old clothes)

{{transferTest|fij|eng|basu|tear down}} (e.g. old buildings)

*Case 4:Fijian does not distinguish genders on pronouns.

{{transferTest|fij|eng|koya|him}}

{{transferTest|fij|eng|koya|her}}

{{transferTest|fij|eng|koya|it}}

*Case 5: (a disambiguation problem?)
The word ''levu'' can be used either as an adjective meaning "big", or a number meaning "many, much", but both numbers and adjectives can be a predicate head (like a verb).

{{transferTest|fij|eng|levu|big}} (adj)

{{transferTest|fij|eng|levu|a lot of}} (num)

==Additions==
*126 more stems in the Fijian transducer and the bilingual dictionary (finished adding all stems to the transducer; continuing adding them to the bilingual dictionary)
*Morphologies:
:-Causative prefix ''vaka-''
:-Collective prefix ''vei-''
:-Deriving verb from noun: prefix "i-"
*Lexical Selection rules:
:-Case 1: Select 'arrive' as the translation of ''yaco'' when the subject NP following the verb is something that can move around; select 'happen' as the translation for ''yaco'' when the subject NP is inanimate.

{{transferTest|fij|eng|yaco|arrive}}

{{transferTest|fij|eng|yaco|happen}}

*Disambiguation rules:

:-Several verbs in Fijian can be used as post-head adverbs, such as ''oti'' ('already' or 'finish').
:*Rules: If it follows a verb or an object pronoun, then choose <adv>; if it follows an aspect/tense marker or a subject pronoun, choose <v>.

:-The word ''soqo'' can be either a verb, meaning 'gather', or a noun, meaning 'meeting'.
:*Rules: If it follows an article, choose <n>; if it follows an aspect/tense marker or a subject pronoun, choose <v>.

==Final Evaluation==
===Fijian Transducer===
*Precision: 92.19219%
*Recall: 55.11670%
*Coverage over the large corpus: 70.06%
*Number of words in the large corpus: 1099762
*Number of stems in the transducer: 280

===MT===
====fij → eng====
*WER:132.31%
*PER:123.26%
*Proportion of correctly translated stems: 16.7%
*Trimmed coverage over <code>fij.longer.text</code>:45.49%
*Trimmed coverage over <code>fij.corpus.large</code>:33.5%
*Number of tokens in <code>longer</code> corpus:1076
====eng → fij====
*WER:92.94%
*PER:85.32%
*Proportion of correctly translated stems:14.68%
*Trimmed coverage of <code>eng.longer.txt</code>:47.67%
*Number of tokens in <code>longer</code> corpus:718.

==Contrastive Grammar==
https://wikis.swarthmore.edu/ling073/Fijian_and_English/Contrastive_Grammar

==Developed Resources for Machine Translation==
https://github.swarthmore.edu/hwang11/ling073-fij-eng

https://github.swarthmore.edu/hwang11/ling073-fij-eng-corpus

[[Category:Sp18_TranslationPairs]]

Fijian/Universal Dependencies

2018-05-05T03:35:25Z

Hwang11: /* Dependency Relations */

==Evaluation==
{| class="wikitable"
!
! fij.withmorph.udpipe
! fij.nomorph.udpipe
! Number of forms
! Number of sentences
|-
! fij.annotated.ud.conllu
| UAS: 91.49%, LAS: 87.59%
| UAS: 83.69%, LAS: 79.79%
| 282
| 30
|-
! fij.annotated2.ud.conllu
| UAS: 55.17%, LAS: 39.08%
| UAS: 50.57%, LAS: 38.51%
| 174
| 10
|}

==Dependency Relations==
===nsubj===
The <code>nsubj</code> denotes the noun phrase or pronoun that is the syntactic subject of the predicate. The subject NP always follow the predicate. When there is no subject NP present, the only pronoun preceding the predicate head, which can be verb, adjective, noun, or even number, gets the <code>nsubj</code>.
*Example 1: The NP ''o Jone'' following the predicate of the subject clause ''la’o'' ('go') is <code>nsubj</code> of the predicate ''la’o''.
<pre>
"<E>"
"e" prn pers p3 sg subj @expl #1->2
"<sega>"
"sega" vaux @root #2->0
"<ni>"
"ni" rel @mark #3->4
"<la’o>"
"la’o" vblex iv @csubj #4->2
"<o>"
"o" art @det #5->6
"<Jone>"
"Jone" np ant @nsubj #6->4
"<.>"
"." sent @punct #7->2
</pre>
'Jone is not going.' (Lit: 'It is not the case that John is going.')
*Example 2: This sentence does not have a subject NP. The pronoun preceding the predicate head ''lesu'' ('return') is the subject of the clause, and thus gets <code>nsubj</code>
<pre>
"<O>"
"o" prn pers p2 sg subj1 @nsubj #1->2
"<lesu>"
"lesu" vblex iv @root #2->0
"<mai>"
"mai" pr @advmod #3->2
"<.>"
"." sent @punct #4->2
</pre>
'You return here.'

===expl===
A predicate head usually requires a subject pronoun preceding it. When there is a subject NP or subject clause with the same reference following the predicate, the subject pronoun gets the <code>expl</code> relation. When no subject NP nor subject clause is present, the subject pronoun preceding the predicate is not an <code>expl</code> but rather a <code>subj</code> as described above. The <code>expl</code> in Fijian corresponds to the second case of expletive relation listed on the UD relations website--true clitic doubling.
*Example 1: The head of the NP ''dra’i'' is the subject of the predicate ''caa'' ('(to be) bad'), and so the pronoun ''e'', referring to the ''a dra’i'' ('weather'), is the <code>expl</code>.
<pre>
"<E>"
"e" prn pers p3 sg subj @expl #1->3
"<rui>"
"rui" adv @advmod #2->3
"<caa>"
"caa" adj @root #3->0
"<a>"
"a" art @det #4->5
"<dra’i>"
"dra’i" n @nsubj #5->3
"<.>"
"." sent @punct #6->3
</pre>
'The weather is unusually bad.'
*Example 2: The cardinal pronouns ''ira'' and ''rau'' are used like focus to contrast the two clauses. The subject pronouns are omitted when cardinal pronouns are present. Since ''a marama'' ('the women') is the subject of the predicate in the first clause, the clitic cardinal pronoun ''ira'' gets the <code>expl</code>; since no subject NP is present in the second clause, the cardinal pronoun ''rau'' is the <code>nsubj</code>.
<pre>
"<La’o>"
"la’o" vblex iv @root #1->0
"<o>"
"o" art @det #2->3
"<ira>"
"ra" prn pers p3 pl card1 @expl #3->1
"<a>"
"a" art @det #4->5
"<marama>"
"marama" n @nsubj #5->1
"<;>"
";" sent @punct #6->7
"<la’o>"
"la’o" vblex iv @conj #7->1
"<mai>"
"mai" pr @advmod #8->7
"<o>"
"o" art @det #9->10
"<rau>"
"rau" prn pers p3 du card2 @nsubj #10->7
"<.>"
"." sent @punct #11->7
</pre>
'They (plural), the women went (home) and they (dual) (i.e. the men) came here.'
===obl===
As described on the UD relations website, the nominal that is not the core argument of the predicate (i.e. subject or direct object) but modifies the predicate is the <code>obl</code> of the predicate. In Fijian, the oblique nominal is usually preceded by a preposition.
*Example 1: The proper name ''’Orovou'' following the preposition ''mai'' ('at') is the <code>obl</code> of the predicate ''ca’a'' ('do').
<pre>
"<E>"
"e" prn pers p3 sg subj @expl #1->2
"<sega>"
"sega" vaux @root #2->0
"<ni>"
"ni" rel @mark #3->5
"<o>"
"o" art @det #4->5
"<Viidawa>"
"Viidawa" np top @csubj #5->2
"<,>"
"," cm @punct #6->8
"<e>"
"e" prn pers p3 sg subj @expl #7->8
"<ca’a>"
"ca’a" vblex iv @conj #8->2
"<mai>"
"mai" pr @case #9->10
"<’Orovou>"
"’Orovou" np top @obl #10->8
"<a>"
"a" art @det #11->12
"<soqo>"
"soqo" n @nsubj #12->8
"<.>"
"." sent @punct #13->8
</pre>
'It's not Viidawa, the meeting is being held at ’Orovou.'
*Example 2:
<pre>
"<’Ua>"
"’ua" vaux @root #1->0
"<ni>"
"ni" rel @mark #2->4
"<la’i>"
"la’i" adv @advmod #3->4
"<taaoo>"
"taaoo" adj @csubj #4->1
"<tale>"
"tale" adv @advmod #5->4
""
"i" pr @case #6->7
"<Viidawa>"
"Viidawa" np top @obl #7->4
"<,>"
"," cm @punct #8->9
"<la’o>"
"la’o" vblex iv @conj #9->1
"<sara>"
"sara" adv @advmod #10->9
""
"i" pr @case #11->12
"<’Orovou>"
"’Orovou" np top @obl #12->9
"<!>"
"!" sent @punct #13->9
</pre>
'Don't get held up at Viidawa, go straight on to ’Orovou.'
===advmod===
The adverb or adverbial phrase modifying the predicate gets the <code>advmod</code> relation. The <code>advmod</code> is not taken by adverbs only; adjectives (e.g. Example 1 below) sometimes can also be <code>advmod</code>. Fijian sentences tend to use a lot of adverbs (or "modifiers" as called in the grammar book) to modify the predicate head, and some of the words may not be directly translated as adverbs.
*Example 1: The adjective ''vina’a'' ('good') modifies the action ''vava'' ('(baby) be carried on back') and thus is the <code>advmod</code>.
<pre>
"<Sa>"
"sa" asp @aux #1->2
"<vava>"
"vava" vblex iv @root #2->0
"<vina’a>"
"vina’a" adj @advmod #3->2
"<a>"
"a" art @det #4->5
"<gone>"
"gone" n @nsubj #5->2
"<yaa>"
"yaa" det dem @det #6->5
"<?>"
"?" sent @punct #7->2
</pre>
'Is the baby being carried on back properly?'
*Example 2: ''sara'' and ''gaa'' are both post head modifiers: ''sara'' means 'very', 'immediately', and 'right on'; ''gaa'' has the meanings 'only', 'particularly', 'just', 'still'. They both modify the
<pre>
"<Sa>"
"sa" asp @aux #1->2
"<marama>"
"marama" n @root #2->0
"<savasavaa>"
"savasavaa" adj @amod #3->2
"<sara>"
"sara" adv @advmod #4->2
"<gaa>"
"gaa" adv @advmod #5->2
"<o>"
"o" art @det #6->7
"<Aneta>"
"Aneta" np ant @nsubj #7->2
"<.>"
"." sent @punct #8->2
</pre>
'Aneta is a very clean lady.'
===xcomp===
The <code>xomp</code> is a subordinate clause with a covert subject coreferent with an argument outside the clause, i.e. the subject or object in the main clause. In Fijian, <code>xcomp</code> is usually the secondary predicate without its own subject following a preposition or relator with the meaning similar to 'to' in English (Example 1) or following a modal verb like "incline", "want", and etc (Example 2).
*Example 1:
<pre>
"<E>"
"e" prn pers p3 sg subj @nsubj #1->3
"<sa>"
"sa" asp @aux #2->3
"<la’o>"
"la’o" vblex iv @root #3->0
""
"i" pr @mark #4->5
"<siwa>"
"siwa" vblex iv @xcomp #5->3
"<.>"
"." sent @punct #6->3
</pre>
'He is going to fish.'
*Example 2:
<pre>
"<Saa>"
"saa" asp @aux #1->2
"<via>"
"via" vbmod @root #2->0
"<levu>"
"levu" adj @xcomp #3->2
"<ca’e>"
"ca’e" adv @advmod #4->3
"<mai>"
"mai" adv @advmod #5->3
"<a>"
"a" art @det #6->7
"<wai>"
"wai" n @nsubj #7->2
"<.>"
"." sent @punct #8->2
</pre>
'The water is rising.' (Lit: 'The water is inclining more towards being big up there.')

==Corpus==
For more relations as well as their examples, please see fij.annotated.ud.txt and fij.annotated2.ud.txt file in the repository below.
https://github.swarthmore.edu/hwang11/ling073-fij-corpus

[[Category: sp18_UD]]

Fijian/Universal Dependencies

2018-05-05T02:21:42Z

2018-05-04T13:14:52Z

Hwang11: /* Developed Resources */

=External Resources=
==Computational Resources==
Spell checker: http://www.spellcheck.net/fiji_spell_checker.html.

==Dictionary==
Fijian-English online dictionary: https://glosbe.com/en/fj.

==Grammar==
Dixon, R.M.W.. 1988. ''A grammar of Boumaa Fijian''. The University of Chicago Press, Chicago.

==Corpora==
Bible translation: http://listen.bible.is/FIJBSP/Matt/1/D.

Wikipedia pages: https://fj.wikipedia.org/wiki/Special:AllPages.

=Developed Resources=
==Keyboard==
https://wikis.swarthmore.edu/ling073/Fijian/Keyboard

https://github.swarthmore.edu/hwang11/ling073-fij-keyboard

==Corpus==
https://github.swarthmore.edu/hwang11/ling073-fij-corpus

==Grammar Documentation==
https://wikis.swarthmore.edu/ling073/Fijian/Grammar.

==Transducer==
https://wikis.swarthmore.edu/ling073/Fijian/Transducer

https://github.swarthmore.edu/hwang11/ling073-fij

==Machine Translation==
https://wikis.swarthmore.edu/ling073/Fijian_and_English

https://wikis.swarthmore.edu/ling073/Fijian_and_English/Contrastive_Grammar

https://github.swarthmore.edu/hwang11/ling073-fij-eng-corpus

==Universal Dependencies==
https://wikis.swarthmore.edu/ling073/Fijian/Universal_Dependencies

[[Category:sp18_ResourceDocumentation]] [[Category: Fijian]]

Fijian/Universal Dependencies

2018-05-04T13:14:06Z

Hwang11:

2018-04-22T02:05:42Z

Hwang11: /* Additions */

Fijian and English

2018-04-22T01:57:27Z

Hwang11: /* Additions */

Fijian and English

2018-04-20T23:52:17Z

Hwang11:

Fijian and English/Structural transfer

2018-04-17T15:23:03Z

Hwang11: /* Post-evaluation */

==Initial Evaluation==
*eng → fij
:-WER:75.81%
:-PER:70.97%
*fij → eng
:-WER:82.14%
:-PER:76.79%

==eng → fij==
I am working on the first contrastive grammar point--word orders within NPs. An English NP has the demonstrative and adjectives preceding the head noun, while Fijian has all of the modifiers following the head noun. Besides, the article in front of the head noun is obligatory even if a demonstrative is present, but its corresponding translation in English--the determiner ''the''--is not allowed to occur when there is a demonstrative in the NP. In terms of tags, Fijian does not have number tags on either nouns or demonstratives.
*Example phrase: {{transferTest|eng|fij|this big village|a ’oro levu yai}}
*Current translation to Fijian:<code>#yai levu #’oro</code>
*Taggers: <code>^this/this<det><dem><sg>$ ^big/big<adj><sint>$ ^village/village<n><sg>$^./.<sent>$</code>
*Biltrans: <code>^this<det><dem><sg>/yai<det><dem><sg>$ ^big<adj><sint>/levu<adj>$ ^village<n><sg>/’oro<n><sg>$^.<sent>/.<sent>$</code>
Step 1: remove number tags and add ''a<art>'' in the nom chunk:
*Chunker: <code>^dem<det><dem>{^yai<det><dem>$}$ ^adj<adj><sint>{^levu<adj>$}$ ^nom<n>{^a<art> ^’oro<n>$}$^sent<SENT>{^.<sent>$}$</code>
Step 2: change word orders:
*Interchunk: <code>^nom<n>{^a<art> ^’oro<n>$}$ ^adj<adj><sint>{^levu<adj>$}$ ^dem<det><dem>{^yai<det><dem>$}$^sent<SENT>{^.<sent>$}$</code>
*Postchunk: <code>^a<art>$ ^’oro<n>$ ^levu<adj>$ ^yai<det><dem>$^.<sent>$</code>
*Translation: <code>a ’oro levu yai</code>

==fij → eng==
===Same grammar point as above===

The relationship between the Fijian article ''a'' and the English definite determiner ''the'' was complicated to implement. For now, I set the article ''a'' to translate to null all the time in the bilingual dictionary. This works here and for most complex NPs such as those with demonstratives and possessives, but not for NPs without any <det>, such as ''a ’oro levu'' (ART village big) or ''a ’oro'' (ART village); for these simple NPs, I will need to write separate rules with the two patterns in the fij-eng.t1x file.
*Example phrase: {{transferTest|fij|eng|a ’oro levu yai|this big village}}
*Current translation from Fijian to English: <code>#village big #this</code>
*Taggers:
:-Fijian:<code>^a/a<art>$ ^’oro/’oro<n>$ ^levu/levu<adj>$ ^yai/yai<det><dem>$^./.<sent>$</code>
:-English(desired):<code>^this/this<det><dem><sg>$ ^big/big<adj><sint>$ ^village/village<n><sg>$^./.<sent>$</code>
*Biltrans:<code>^a<art>/$ ^’oro<n>/village<n>$ ^levu<adj>/big<adj><sint>$ ^yai<det><dem>/this<det><dem>$^.<sent>/.<sent>$</code>
Fijian nouns and demonstratives do not distinguish numbers, so I want to add a number tag for the English noun and the demonstrative ''this'', and then change the word order from head first in Fijian to head final in English:
<pre>
<rule comment="REGLA: NPs with demonstratives">
<pattern>
<pattern-item n="article"/>
<pattern-item n="noun_reg"/>
<pattern-item n="adj"/>
<pattern-item n="dem"/>
</pattern>
<action>
<out>
<chunk name="nom_adj_dem" case="caseFirstWord">
<tags>
<tag><lit-tag v="SN"/></tag>
<tag><lit-tag v="sg"/></tag>
</tags>
<lu>
<clip pos="4" side="tl" part="lem"/>
<clip pos="4" side="tl" part="a_pos"/>
<clip pos="4" side="tl" part="subcategories"/>
<lit-tag v="sg"/>
</lu>

<lu>
<clip pos="3" side="tl" part="whole"/>
</lu>

<lu>
<clip pos="2" side="tl" part="lem"/>
<clip pos="2" side="tl" part="a_pos"/>
<lit-tag v="sg"/>
</lu>
</chunk>
</out>
</action>
</rule>
</pre>
*Chunker:<code>^nom_adj_dem<SN><sg>{^this<det><dem><sg>$ ^big<adj><sint>$ ^village<n><sg>$}$^sent<SENT>{^.<sent>$}$</code>
*Interchunk:<code>^nom_adj_dem<SN><sg>{^this<det><dem><sg>$ ^big<adj><sint>$ ^village<n><sg>$}$^sent<SENT>{^.<sent>$}$</code>
*Postchunk:<code>^this<det><dem><sg>$ ^big<adj><sint>$ ^village<n><sg>$^.<sent>$</code>
*Note and problems with current codes:
Before this I tried to add number tags by having <code><clip pos="2" side="tl" part="a_num"></code> in the lexical unit section, but since Fijian nouns do not change for numbers (e.g. ’oro can be translated as both 'village' and 'villages'), the English noun still did not what number tag to take and thus the number tag never showed up in the chunk. Instead, I added the number tag manually to the noun and demonstrative in the English translation by <lit-tag>, setting it to be "sg" in this pattern. It seems to work in this case, but in fact without contexts from sentences, it is inaccurate to translate "a ’oro levu yai" as "this big village" (e.g. {{transferTest|fij|eng|Au raica (koya) a ’oro levu yai|I see this big village.}} and {{transferTest|fij|eng|Au raica ira a ’oro levu yai|I see these big villages.}}). So I need to change my example from a phrase to a sentence with enough contexts and write a rule for that pattern instead.
===In sentences===
*Example sentence: {{transferTest|fij|eng|Au raici ira a gone.|I see the children.}}
*Codes:
<pre>
<rule comment="REGLA: Clause_1">
<pattern>
<pattern-item n="pronoun"/>
<pattern-item n="tense_mk"/>
<pattern-item n="verb"/>
<pattern-item n="pronoun"/>
<pattern-item n="article"/>
<pattern-item n="noun_reg"/>
</pattern>
<action>
<let>
<var n="tense"/>
<clip pos="2" side="sl" part="a_tense"/>
</let>
<out>
<chunk name="Clause_1" case="caseFirstWord">
<tags>
<tag><lit-tag v="SV"/></tag>
<tag><var n="tense"/></tag>
<tag><lit-tag v="SN"/></tag>
<tag><lit-tag v="pl"/></tag>
</tags>
<lu>
<clip pos="1" side="tl" part="whole"/>
</lu>

<lu>
<clip pos="3" side="tl" part="lem"/>
<clip pos="3" side="tl" part="a_pos"/>
<clip pos="2" side="sl" part="a_tense"/>
</lu>

<lu>
<lit v="the"/>
<lit-tag v="det.def.sp"/>
</lu>

<lu>
<clip pos="6" side="tl" part="lem"/>
<clip pos="6" side="tl" part="a_pos"/>
<clip pos="4" side="sl" part="a_num"/>
</lu>
</chunk>
</out>
</action>
</rule>
</pre>
*Chunker: <code>^clause_2<SV><pres><SN><pl>{^Prpers<prn><p1><mf><sg><subj>$ ^see<vblex><pres>$ ^the<det><def><sp>$ ^child<n><pl>$}$^sent<SENT>{^.<sent>$}$^sent<SENT>{^.<sent>$}$</code>
*Interchunk: <code>^clause_2<SV><pres><SN><pl>{^Prpers<prn><p1><mf><sg><subj>$ ^see<vblex><pres>$ ^the<det><def><sp>$ ^child<n><pl>$}$^sent<SENT>{^.<sent>$}$^sent<SENT>{^.<sent>$}$</code>
*Postchunk: <code>^Prpers<prn><p1><mf><sg><subj>$ ^see<vblex><pres>$ ^the<det><def><sp>$ ^child<n><pl>$^.<sent>$^.<sent>$</code>
*Translation: I see the children.

==Post-evaluation==
*eng → fij
:-WER: 64.52%
:-PER: 56.45%
*fij → eng
:-WER: 51.79%
:-PER: 46.43%

Fijian and English/Structural transfer

2018-04-17T15:21:18Z

Hwang11: /* fij → eng */

==Initial Evaluation==
*eng → fij
:-WER:75.81%
:-PER:70.97%
*fij → eng
:-WER:82.14%
:-PER:76.79%

==eng → fij==
I am working on the first contrastive grammar point--word orders within NPs. An English NP has the demonstrative and adjectives preceding the head noun, while Fijian has all of the modifiers following the head noun. Besides, the article in front of the head noun is obligatory even if a demonstrative is present, but its corresponding translation in English--the determiner ''the''--is not allowed to occur when there is a demonstrative in the NP. In terms of tags, Fijian does not have number tags on either nouns or demonstratives.
*Example phrase: {{transferTest|eng|fij|this big village|a ’oro levu yai}}
*Current translation to Fijian:<code>#yai levu #’oro</code>
*Taggers: <code>^this/this<det><dem><sg>$ ^big/big<adj><sint>$ ^village/village<n><sg>$^./.<sent>$</code>
*Biltrans: <code>^this<det><dem><sg>/yai<det><dem><sg>$ ^big<adj><sint>/levu<adj>$ ^village<n><sg>/’oro<n><sg>$^.<sent>/.<sent>$</code>
Step 1: remove number tags and add ''a<art>'' in the nom chunk:
*Chunker: <code>^dem<det><dem>{^yai<det><dem>$}$ ^adj<adj><sint>{^levu<adj>$}$ ^nom<n>{^a<art> ^’oro<n>$}$^sent<SENT>{^.<sent>$}$</code>
Step 2: change word orders:
*Interchunk: <code>^nom<n>{^a<art> ^’oro<n>$}$ ^adj<adj><sint>{^levu<adj>$}$ ^dem<det><dem>{^yai<det><dem>$}$^sent<SENT>{^.<sent>$}$</code>
*Postchunk: <code>^a<art>$ ^’oro<n>$ ^levu<adj>$ ^yai<det><dem>$^.<sent>$</code>
*Translation: <code>a ’oro levu yai</code>

==fij → eng==
===Same grammar point as above===

The relationship between the Fijian article ''a'' and the English definite determiner ''the'' was complicated to implement. For now, I set the article ''a'' to translate to null all the time in the bilingual dictionary. This works here and for most complex NPs such as those with demonstratives and possessives, but not for NPs without any <det>, such as ''a ’oro levu'' (ART village big) or ''a ’oro'' (ART village); for these simple NPs, I will need to write separate rules with the two patterns in the fij-eng.t1x file.
*Example phrase: {{transferTest|fij|eng|a ’oro levu yai|this big village}}
*Current translation from Fijian to English: <code>#village big #this</code>
*Taggers:
:-Fijian:<code>^a/a<art>$ ^’oro/’oro<n>$ ^levu/levu<adj>$ ^yai/yai<det><dem>$^./.<sent>$</code>
:-English(desired):<code>^this/this<det><dem><sg>$ ^big/big<adj><sint>$ ^village/village<n><sg>$^./.<sent>$</code>
*Biltrans:<code>^a<art>/$ ^’oro<n>/village<n>$ ^levu<adj>/big<adj><sint>$ ^yai<det><dem>/this<det><dem>$^.<sent>/.<sent>$</code>
Fijian nouns and demonstratives do not distinguish numbers, so I want to add a number tag for the English noun and the demonstrative ''this'', and then change the word order from head first in Fijian to head final in English:
<pre>
<rule comment="REGLA: NPs with demonstratives">
<pattern>
<pattern-item n="article"/>
<pattern-item n="noun_reg"/>
<pattern-item n="adj"/>
<pattern-item n="dem"/>
</pattern>
<action>
<out>
<chunk name="nom_adj_dem" case="caseFirstWord">
<tags>
<tag><lit-tag v="SN"/></tag>
<tag><lit-tag v="sg"/></tag>
</tags>
<lu>
<clip pos="4" side="tl" part="lem"/>
<clip pos="4" side="tl" part="a_pos"/>
<clip pos="4" side="tl" part="subcategories"/>
<lit-tag v="sg"/>
</lu>

<lu>
<clip pos="3" side="tl" part="whole"/>
</lu>

<lu>
<clip pos="2" side="tl" part="lem"/>
<clip pos="2" side="tl" part="a_pos"/>
<lit-tag v="sg"/>
</lu>
</chunk>
</out>
</action>
</rule>
</pre>
*Chunker:<code>^nom_adj_dem<SN><sg>{^this<det><dem><sg>$ ^big<adj><sint>$ ^village<n><sg>$}$^sent<SENT>{^.<sent>$}$</code>
*Interchunk:<code>^nom_adj_dem<SN><sg>{^this<det><dem><sg>$ ^big<adj><sint>$ ^village<n><sg>$}$^sent<SENT>{^.<sent>$}$</code>
*Postchunk:<code>^this<det><dem><sg>$ ^big<adj><sint>$ ^village<n><sg>$^.<sent>$</code>
*Note and problems with current codes:
Before this I tried to add number tags by having <code><clip pos="2" side="tl" part="a_num"></code> in the lexical unit section, but since Fijian nouns do not change for numbers (e.g. ’oro can be translated as both 'village' and 'villages'), the English noun still did not what number tag to take and thus the number tag never showed up in the chunk. Instead, I added the number tag manually to the noun and demonstrative in the English translation by <lit-tag>, setting it to be "sg" in this pattern. It seems to work in this case, but in fact without contexts from sentences, it is inaccurate to translate "a ’oro levu yai" as "this big village" (e.g. {{transferTest|fij|eng|Au raica (koya) a ’oro levu yai|I see this big village.}} and {{transferTest|fij|eng|Au raica ira a ’oro levu yai|I see these big villages.}}). So I need to change my example from a phrase to a sentence with enough contexts and write a rule for that pattern instead.
===In sentences===
*Example sentence: {{transferTest|fij|eng|Au raici ira a gone.|I see the children.}}
*Codes:
<pre>
<rule comment="REGLA: Clause_1">
<pattern>
<pattern-item n="pronoun"/>
<pattern-item n="tense_mk"/>
<pattern-item n="verb"/>
<pattern-item n="pronoun"/>
<pattern-item n="article"/>
<pattern-item n="noun_reg"/>
</pattern>
<action>
<let>
<var n="tense"/>
<clip pos="2" side="sl" part="a_tense"/>
</let>
<out>
<chunk name="Clause_1" case="caseFirstWord">
<tags>
<tag><lit-tag v="SV"/></tag>
<tag><var n="tense"/></tag>
<tag><lit-tag v="SN"/></tag>
<tag><lit-tag v="pl"/></tag>
</tags>
<lu>
<clip pos="1" side="tl" part="whole"/>
</lu>

<lu>
<clip pos="3" side="tl" part="lem"/>
<clip pos="3" side="tl" part="a_pos"/>
<clip pos="2" side="sl" part="a_tense"/>
</lu>

<lu>
<lit v="the"/>
<lit-tag v="det.def.sp"/>
</lu>

<lu>
<clip pos="6" side="tl" part="lem"/>
<clip pos="6" side="tl" part="a_pos"/>
<clip pos="4" side="sl" part="a_num"/>
</lu>
</chunk>
</out>
</action>
</rule>
</pre>
*Chunker: <code>^clause_2<SV><pres><SN><pl>{^Prpers<prn><p1><mf><sg><subj>$ ^see<vblex><pres>$ ^the<det><def><sp>$ ^child<n><pl>$}$^sent<SENT>{^.<sent>$}$^sent<SENT>{^.<sent>$}$</code>
*Interchunk: <code>^clause_2<SV><pres><SN><pl>{^Prpers<prn><p1><mf><sg><subj>$ ^see<vblex><pres>$ ^the<det><def><sp>$ ^child<n><pl>$}$^sent<SENT>{^.<sent>$}$^sent<SENT>{^.<sent>$}$</code>
*Postchunk: <code>^Prpers<prn><p1><mf><sg><subj>$ ^see<vblex><pres>$ ^the<det><def><sp>$ ^child<n><pl>$^.<sent>$^.<sent>$</code>
*Translation: I see the children.

==Post-evaluation==
*eng → fij
:-WER: 66.13%
:-PER: 61.29%
*fij → eng
:-WER: 75.00%
:-PER: 75.00%

Fijian and English/Contrastive Grammar

2018-04-17T14:16:51Z

Hwang11:

=Fij-Eng Tests=
==Noun Phrase Structure==
===Word orders===
In a Fijian NP, adjectives, demonstratives, relative clauses modifying the head noun, if present, must follow the head noun, while in English, adjectives and demonstratives precede the head noun and relative clauses follow the head noun.

*Example 1:

{{transferTest|fij|eng|a ’oro levu yai|this big village}} (gloss: ART village big this)

{{transferMorphTest|fij|eng|a{{tag|art}} ’oro{{tag|n}} levu{{tag|adj}} yai{{tag|det}}{{tag|dem}}|this{{tag|det}}{{tag|dem}}{{tag|sg}} big{{tag|adj}}{{tag|sint}} village{{tag|n}}{{tag|sg}}}}

*Example 2:

{{transferTest|fij|eng|a gone vina’a yaa|that good child}} (gloss: ART child good that)

{{transferMorphTest|fij|eng|a{{tag|art}} gone{{tag|n}} vina’a{{tag|adj}} yaa{{tag|det}}{{tag|dem}}|that{{tag|det}}{{tag|dem}}{{tag|sg}} good{{tag|adj}}{{tag|sint}} child{{tag|n}}{{tag|sg}}}}

===Nominal articles===
An NP in Fijian has to begin with an article: ''a'' for common nouns (which becomes ''na'' after a preposition) and ''o'' for proper names and pronouns (which is dropped after a preposition). The articles in Fijian roughly corresponds to the definite determiner ''the'' in English, and there's no indefinite article in Fijian. The indefinite article in English can be translated through the number "one".

*Example 1.1:

{{transferTest|fij|eng|a ’oro|the village}} (gloss: ART village)

{{transferMorphTest|fij|eng|a{{tag|art}} ’oro{{tag|n}}|the{{tag|det}}{{tag|def}}{{tag|sp}} village{{tag|n}}{{tag|sg}}}}

*Example 1.2:

{{transferTest|fij|eng|e dua a ’oro|a village}} (gloss: 3sg-subj one ART village)

{{transferMorphTest|fij|eng|e{{tag|prn}}{{tag|pers}}{{tag|p3}}{{tag|sg}}{{tag|subj}} dua{{tag|num}} a{{tag|art}} ’oro{{tag|n}}|a{{tag|det}}{{tag|ind}}{{tag|sg}} village{{tag|n}}{{tag|sg}}}}

*Example 2.1:
{{transferTest|fij|eng|o Viti|Fiji}} (gloss: ART Suva)

{{transferMorphTest|fij|eng|o{{tag|art}} Viti{{tag|np}}{{tag|top}}|Fiji{{tag|np}}{{tag|top}}{{tag|sg}}}}

{{comment|could you give some examples of some of the other patterns you mentioned? in theory each point should have multiple examples, and this one needs it the most}}

===Possessive structures===
When the possessor is a pronoun and the possessed is a bound noun, the bound noun is marked with a possessive suffix in Fijian.
*Example 1:

{{transferTest|fij|eng|a mataqu|my eye}} (gloss: ART eye-1sg)

{{transferMorphTest|fij|eng|a{{tag|art}} mata{{tag|n}}{{tag|pos}}{{tag|p1}}{{tag|sg}}|my{{tag|det}}{{tag|pos}}{{tag|p1}}{{tag|mf}}{{tag|sg}} eye{{tag|n}}{{tag|sg}}}}

*Example 2:

{{transferTest|fij|eng|a tamamu|your father}} (gloss: ART father-2sg)

{{transferMorphTest|fij|eng|a{{tag|art}} tama{{tag|n}}{{tag|pos}}{{tag|p2}}{{tag|sg}}|your{{tag|det}}{{tag|pos}}{{tag|p2}}{{tag|mf}}{{tag|sp}} father{{tag|n}}{{tag|sg}}}}

{{comment|it might be better to just say that the noun is marked with possession morphology in Fijian (if I understand correctly?) and an article occurs with it}}

*Note: Fijian distinguishes between bound nouns, which must be followed by one of the following: a pronominal suffix; ''-i'' plus name; or ''ni'' plus NP, and free nouns, which can stand alone. Bound nouns generally include kin terms used to reference to the kins, primary body parts (e.g. ''ulu-'' "head", ''yava-'' "leg, foot", ''mata-'' "face", ''sui-'' "bone", ''’uli-'' "skin" etc., but not ''draa'' "blood"), ''yalo-'' "spirit", ''yaca-'' "name", and several nouns referring to abstract qualities of things (e.g. ti’i- "piece, portion of", vu’u- "cause of", etc.).

==Predicate Structure==
===Word orders===
The major clause type in Fijian always has a predicate at the beginning of the clause, usually with a verb as the head and a obligatory subject pronoun in the initial position of the predicate. If a subject NP is present (which is optional), it must follow the predicate. The object pronoun and object NP with the same reference for a transitive verb follow the verb and stand within the predicate. An English clause generally has an SVO word order, and a pronoun and a full NP with the same reference cannot co-occur as the subject of a verb.
====Intransitive====

*Example 1.1: When there's no subject NP, the surface word order in Fijian is the same as that in English.

{{transferTest|fij|eng|Au rai.|I look.}} (gloss: 1sg-subj look)

{{transferMorphTest|fij|eng|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} rai{{tag|vblex}}{{tag|iv}} .{{tag|sent}}|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} look{{tag|vblex}}{{tag|pres}} .{{tag|sent}}}}

*Example 1.2:

{{transferTest|fij|eng|Era la’o a gone.|The children go.}} (gloss: 3pl-subj go ART child)

{{transferMorphTest|fij|eng|ra{{tag|prn}}{{tag|pers}}{{tag|p3}}{{tag|pl}}{{tag|subj1}} la’o{{tag|vblex}}{{tag|iv}} a{{tag|art}} gone{{tag|n}} .{{tag|sent}}|the{{tag|det}}{{tag|def}}{{tag|sp}} child{{tag|n}}{{tag|pl}} go{{tag|vblex}}{{tag|pres}} .{{tag|sent}}}}

====Transitive====
*Example 2.0: When the object is a 3sg pronoun, the following sentence is acceptable in Fijian, but not in English without the object explicit.

{{transferTest|fij|eng|Au raica.|I see him.}} (Lit: *I see.)

{{transferMorphTest|fij|eng|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} rai{{tag|vblex}}{{tag|tv}} .{{tag|sent}}|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} see{{tag|vblex}}{{tag|pres}} prpers{{tag|prn}}{{tag|p3}}{{tag|m}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}}}

{{transferTest|fij|eng|Au raica.|I see her.}}

{{transferMorphTest|fij|eng|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} rai{{tag|vblex}}{{tag|tv}} .{{tag|sent}}|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} see{{tag|vblex}}{{tag|pres}} prpers{{tag|prn}}{{tag|p3}}{{tag|f}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}}}

{{transferTest|fij|eng|Au raica.|I see it.}}

{{transferMorphTest|fij|eng|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} rai{{tag|vblex}}{{tag|tv}} .{{tag|sent}}|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} see{{tag|vblex}}{{tag|pres}} prpers{{tag|prn}}{{tag|p3}}{{tag|nt}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}}}

*Example 2.1:
{{transferTest|fij|eng|Au raici koya.|I see him.}} (gloss: I see 3sg-obj)

{{transferMorphTest|fij|eng|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} rai{{tag|vblex}}{{tag|tvprn}} koya{{tag|prn}}{{tag|pers}}{{tag|p3}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} see{{tag|vblex}}{{tag|pres}} prpers{{tag|prn}}{{tag|p3}}{{tag|m}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}}}

{{transferTest|fij|eng|Au raici koya.|I see her.}}

{{transferMorphTest|fij|eng|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} rai{{tag|vblex}}{{tag|tvprn}} koya{{tag|prn}}{{tag|pers}}{{tag|p3}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} see{{tag|vblex}}{{tag|pres}} prpers{{tag|prn}}{{tag|p3}}{{tag|f}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}}}

{{transferTest|fij|eng|Au raici koya.|I see it.}}

{{transferMorphTest|fij|eng|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} rai{{tag|vblex}}{{tag|tvprn}} koya{{tag|prn}}{{tag|pers}}{{tag|p3}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} see{{tag|vblex}}{{tag|pres}} prpers{{tag|prn}}{{tag|p3}}{{tag|nt}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}}}

(need disambiguating it)

*Example 2.2:
{{transferTest|fij|eng|Au raici ira a gone.|I see the children.}} (gloss: 1sg-subj see 3pl.obj ART child)

{{transferMorphTest|fij|eng|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} rai{{tag|vblex}}{{tag|tvprn}} ra{{tag|prn}}{{tag|pers}}{{tag|p3}}{{tag|pl}}{{tag|obj}} a{{tag|art}} gone{{tag|n}} .{{tag|sent}}|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} see{{tag|vblex}}{{tag|pres}} the{{tag|det}}{{tag|def}}{{tag|sp}} child{{tag|n}}{{tag|pl}} .{{tag|sent}}}}

====Ambiguity====
The following sentence in Fijian is ambiguous.

{{transferTest|fij|eng|E raica a gone.|He sees the child.}}

{{transferMorphTest|fij|eng|e{{tag|prn}}{{tag|pers}}{{tag|p3}}{{tag|sg}}{{tag|subj}} rai{{tag|vblex}}{{tag|tv}} a{{tag|art}} gone{{tag|n}} .{{tag|sent}}|prpers{{tag|prn}}{{tag|p3}}{{tag|m}}{{tag|sg}}{{tag|subj}} see{{tag|vblex}}{{tag|pres}}{{tag|p3}}{{tag|sg}} the{{tag|det}}{{tag|def}}{{tag|sp}} child{{tag|n}}{{tag|sg}} .{{tag|sent}}}}

{{transferTest|fij|eng|E raica a gone.|The child sees him.}}

{{transferMorphTest|fij|eng|e{{tag|prn}}{{tag|pers}}{{tag|sg}}{{tag|p3}}{{tag|subj}} rai{{tag|vblex}}{{tag|tv}} a{{tag|art}} gone{{tag|n}}{{tag|sg}} .{{tag|sent}}|the{{tag|det}}{{tag|def}}{{tag|sp}} child{{tag|n}} see{{tag|vblex}}{{tag|pres}}{{tag|p3}}{{tag|sg}} prpers{{tag|prn}}{{tag|p3}}{{tag|m}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}}}

{{comment|could you give some examples of some other orders?}}

===Tense-aspect markers===
Fijian uses separate words (known as tense/aspect markers) to mark tenses and aspects, instead of having verb inflections as English does.

{{transferTest|fij|eng|Au aa la’o.|I went.}} (gloss: 1sg-subj PAST go)

{{transferMorphTest|fij|eng|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} aa{{tag|past}} la’o{{tag|vblex}}{{tag|iv}} .{{tag|sent}}|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} go{{tag|vblex}}{{tag|past}} .{{tag|sent}}}}

{{transferTest|fij|eng|Au aa raica a gone.|I saw the child.}}

{{transferMorphTest|fij|eng|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} aa{{tag|past}} rai{{tag|vblex}}{{tag|tv}} a{{tag|art}} gone{{tag|n}} .{{tag|sent}}|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} see{{tag|vblex}}{{tag|past}} the{{tag|det}}{{tag|def}}{{tag|sp}} child{{tag|n}}{{tag|sg}} .{{tag|sent}}}}

(need to disambiguate saw)

{{comment|what do your morphology tags in Fijian look like? This will determine how you characterise the difference.}}

=Eng-Fij Tests=
==Noun Phrase Structure==
===Word orders===
*Example 1:

{{transferTest|eng|fij|this big village|a ’oro levu yai}} (gloss: ART village big this)

{{transferMorphTest|eng|fij|this{{tag|det}}{{tag|dem}}{{tag|sg}} big{{tag|adj}}{{tag|sint}} village{{tag|n}}{{tag|sg}}|a{{tag|art}} ’oro{{tag|n}} levu{{tag|adj}} yai{{tag|det}}{{tag|dem}}}}

*Example 2:

{{transferTest|eng|fij|that good child|a gone vina’a yaa}} (gloss: ART child good that)

{{transferMorphTest|eng|fij|that{{tag|det}}{{tag|dem}}{{tag|sg}} good{{tag|adj}}{{tag|sint}} child{{tag|n}}{{tag|sg}}|a{{tag|art}} gone{{tag|n}} vina’a{{tag|adj}} yaa{{tag|det}}{{tag|dem}}}}

===Nominal articles===
*Example 1.1:

{{transferTest|eng|fij|the village|a ’oro}} (gloss: ART village)

{{transferMorphTest|eng|fij|the{{tag|det}}{{tag|def}}{{tag|sp}} village{{tag|n}}{{tag|sg}}|a{{tag|art}} ’oro{{tag|n}}}}

*Example 1.2:

{{transferTest|eng|fij|a village|e dua a ’oro}} (gloss: 3sg-subj one ART village)

{{transferMorphTest|eng|fij|a{{tag|det}}{{tag|ind}}{{tag|sg}} village{{tag|n}}{{tag|sg}}|e{{tag|prn}}{{tag|pers}}{{tag|p3}}{{tag|sg}}{{tag|subj}} dua{{tag|num}} a{{tag|art}} ’oro{{tag|n}}}}

*Example 2.1:
{{transferTest|eng|fij|Fiji|o Viti}} (gloss: ART Suva)

{{transferMorphTest|eng|fij|Fiji{{tag|np}}{{tag|top}}{{tag|sg}}|o{{tag|art}} Viti{{tag|np}}{{tag|top}}}}

===Poessive structures===
*Example 1:

{{transferTest|eng|fij|my eye|a mataqu}} (gloss: ART eye-1sg)

{{transferMorphTest|eng|fij|my{{tag|det}}{{tag|pos}}{{tag|p1}}{{tag|mf}}{{tag|sg}} eye{{tag|n}}{{tag|sg}}|a{{tag|art}} mata{{tag|n}}{{tag|pos}}{{tag|p1}}{{tag|sg}}}}

*Example 2:

{{transferTest|eng|fij|your father|a tamamu}} (gloss: ART father-2sg)

{{transferMorphTest|eng|fij|your{{tag|det}}{{tag|pos}}{{tag|p2}}{{tag|mf}}{{tag|sp}} father{{tag|n}}{{tag|sg}}|a{{tag|art}} tama{{tag|n}}{{tag|pos}}{{tag|p2}}{{tag|sg}}}}

==Predicate Structure==
===Word orders===
====Intransitive====
*Example 1.1:

{{transferTest|eng|fij|I look.|Au rai.}} (gloss: 1sg-subj look)

{{transferMorphTest|eng|fij|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} look{{tag|vblex}}{{tag|pres}} .{{tag|sent}}|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} rai{{tag|vblex}}{{tag|iv}} .{{tag|sent}}}}

*Example 1.2:

{{transferTest|eng|fij|The children go.|Era la’o a gone.}} (gloss: 3pl-subj go ART child)

{{transferMorphTest|eng|fij|the{{tag|det}}{{tag|def}}{{tag|sp}} child{{tag|n}}{{tag|pl}} go{{tag|vblex}}{{tag|pres}} .{{tag|sent}}|ra{{tag|prn}}{{tag|pers}}{{tag|p3}}{{tag|pl}}{{tag|subj1}} la’o{{tag|vblex}}{{tag|iv}} a{{tag|art}} gone{{tag|n}} .{{tag|sent}}}}

====Transitive====
When the object is a 3sg pronoun, the following English sentence can be translated into Fijian in two ways (ex. 2.0 and 2.1).

*Example 2.0

{{transferTest|eng|fij|I see him.|Au raica.}} (Lit: *I see.)

{{transferMorphTest|eng|fij|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} see{{tag|vblex}}{{tag|pres}} prpers{{tag|prn}}{{tag|p3}}{{tag|m}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} rai{{tag|vblex}}{{tag|tv}} .{{tag|sent}}}}

{{transferTest|eng|fij|I see her.|Au raica.}}

{{transferMorphTest|eng|fij|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} see{{tag|vblex}}{{tag|pres}} prpers{{tag|prn}}{{tag|p3}}{{tag|f}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} rai{{tag|vblex}}{{tag|tv}} .{{tag|sent}}}}

{{transferTest|eng|fij|I see it.|Au raica.}}

{{transferMorphTest|eng|fij|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} see{{tag|vblex}}{{tag|pres}} prpers{{tag|prn}}{{tag|p3}}{{tag|nt}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} rai{{tag|vblex}}{{tag|tv}} .{{tag|sent}}}}

*Example 2.1:
{{transferTest|eng|fij|I see him.|Au raici koya.}} (gloss: I see 3sg-obj)

{{transferMorphTest|eng|fij|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} see{{tag|vblex}}{{tag|pres}} prpers{{tag|prn}}{{tag|p3}}{{tag|m}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} rai{{tag|vblex}}{{tag|tvprn}} koya{{tag|prn}}{{tag|pers}}{{tag|p3}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}}}

{{transferTest|eng|fij|I see her.|Au raici koya.}} (gloss: I see 3sg-obj)

{{transferMorphTest|eng|fij|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} see{{tag|vblex}}{{tag|pres}} prpers{{tag|prn}}{{tag|p3}}{{tag|f}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} rai{{tag|vblex}}{{tag|tvprn}} koya{{tag|prn}}{{tag|pers}}{{tag|p3}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}}}

{{transferTest|eng|fij|I see it.|Au raici koya.}} (gloss: I see 3sg-obj)

{{transferMorphTest|eng|fij|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} see{{tag|vblex}}{{tag|pres}} prpers{{tag|prn}}{{tag|p3}}{{tag|nt}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} rai{{tag|vblex}}{{tag|tvprn}} koya{{tag|prn}}{{tag|pers}}{{tag|p3}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}}}

*Example 2.2:
{{transferTest|eng|fij|I see the children.|Au raici ira a gone.}} (gloss: 1sg-subj see 3pl.obj ART child)

{{transferMorphTest|eng|fij|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} see{{tag|vblex}}{{tag|pres}} the{{tag|det}}{{tag|def}}{{tag|sp}} child{{tag|n}}{{tag|pl}} .{{tag|sent}}|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} rai{{tag|vblex}}{{tag|tvprn}} ra{{tag|prn}}{{tag|pers}}{{tag|p3}}{{tag|pl}}{{tag|obj}} a{{tag|art}} gone{{tag|n}} .{{tag|sent}}}}

====Ambiguity====
The following sentence in Fijian is ambiguous.

{{transferTest|eng|fij|He sees the child.|E raica a gone.}}

{{transferMorphTest|eng|fij|prpers{{tag|prn}}{{tag|p3}}{{tag|m}}{{tag|sg}}{{tag|subj}} see{{tag|vblex}}{{tag|pres}}{{tag|p3}}{{tag|sg}} the{{tag|det}}{{tag|def}}{{tag|sp}} child{{tag|n}}{{tag|sg}} .{{tag|sent}}|e{{tag|prn}}{{tag|pers}}{{tag|p3}}{{tag|sg}}{{tag|subj}} rai{{tag|vblex}}{{tag|tv}} a{{tag|art}} gone{{tag|n}} .{{tag|sent}}}}

{{transferTest|eng|fij|The child sees him.|E raica a gone.}}

{{transferMorphTest|eng|fij|the{{tag|det}}{{tag|def}}{{tag|sp}} child{{tag|n}}{{tag|sg}} see{{tag|vblex}}{{tag|pres}}{{tag|p3}}{{tag|sg}} prpers{{tag|prn}}{{tag|p3}}{{tag|m}}{{tag|sg}}{{tag|obj}} .{{tag|sent}}|e{{tag|prn}}{{tag|pers}}{{tag|sg}}{{tag|p3}}{{tag|subj}} rai{{tag|vblex}}{{tag|tv}} a{{tag|art}} gone{{tag|n}} .{{tag|sent}}}}

===Tense-aspect markers===
*Example 1:

{{transferTest|eng|fij|I went.|Au aa la’o.}} (gloss: 1sg-subj PAST go)

{{transferMorphTest|eng|fij|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} go{{tag|vblex}}{{tag|past}} .{{tag|sent}}|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} aa{{tag|past}} la’o{{tag|vblex}}{{tag|iv}} .{{tag|sent}}}}

*Example 2:

{{transferTest|eng|fij|I saw the child.|Au aa raica a gone.}}

{{transferMorphTest|eng|fij|prpers{{tag|prn}}{{tag|p1}}{{tag|mf}}{{tag|sg}}{{tag|subj}} see{{tag|vblex}}{{tag|past}} the{{tag|det}}{{tag|def}}{{tag|sp}} child{{tag|n}}{{tag|sg}} .{{tag|sent}}|au{{tag|prn}}{{tag|pers}}{{tag|p1}}{{tag|sg}}{{tag|subj2}} aa{{tag|past}} rai{{tag|vblex}}{{tag|tv}} a{{tag|art}} gone{{tag|n}} .{{tag|sent}}}}

[[Category:Sp18_ContrastiveGrammars]]

Fijian and English/Structural transfer

2018-04-17T13:47:46Z

Hwang11: /* fij → eng */

==Initial Evaluation==
*eng → fij
:-WER:75.81%
:-PER:70.97%
*fij → eng
:-WER:82.14%
:-PER:76.79%

==eng → fij==
I am working on the first contrastive grammar point--word orders within NPs. An English NP has the demonstrative and adjectives preceding the head noun, while Fijian has all of the modifiers following the head noun. Besides, the article in front of the head noun is obligatory even if a demonstrative is present, but its corresponding translation in English--the determiner ''the''--is not allowed to occur when there is a demonstrative in the NP. In terms of tags, Fijian does not have number tags on either nouns or demonstratives.
*Example phrase: {{transferTest|eng|fij|this big village|a ’oro levu yai}}
*Current translation to Fijian:<code>#yai levu #’oro</code>
*Taggers: <code>^this/this<det><dem><sg>$ ^big/big<adj><sint>$ ^village/village<n><sg>$^./.<sent>$</code>
*Biltrans: <code>^this<det><dem><sg>/yai<det><dem><sg>$ ^big<adj><sint>/levu<adj>$ ^village<n><sg>/’oro<n><sg>$^.<sent>/.<sent>$</code>
Step 1: remove number tags and add ''a<art>'' in the nom chunk:
*Chunker: <code>^dem<det><dem>{^yai<det><dem>$}$ ^adj<adj><sint>{^levu<adj>$}$ ^nom<n>{^a<art> ^’oro<n>$}$^sent<SENT>{^.<sent>$}$</code>
Step 2: change word orders:
*Interchunk: <code>^nom<n>{^a<art> ^’oro<n>$}$ ^adj<adj><sint>{^levu<adj>$}$ ^dem<det><dem>{^yai<det><dem>$}$^sent<SENT>{^.<sent>$}$</code>
*Postchunk: <code>^a<art>$ ^’oro<n>$ ^levu<adj>$ ^yai<det><dem>$^.<sent>$</code>
*Translation: <code>a ’oro levu yai</code>

==fij → eng==
Same grammar point as above.

The relationship between the Fijian article ''a'' and the English definite determiner ''the'' was complicated to implement. For now, I set the article ''a'' to translate to null all the time in the bilingual dictionary. This works here and for most complex NPs such as those with demonstratives and possessives, but not for NPs without any <det>, such as ''a ’oro levu'' (ART village big) or ''a ’oro'' (ART village); for these simple NPs, I will need to write separate rules with the two patterns in the fij-eng.t1x file.
*Example phrase: {{transferTest|fij|eng|a ’oro levu yai|this big village}}
*Current translation from Fijian to English: <code>#village big #this</code>
*Taggers:
:-Fijian:<code>^a/a<art>$ ^’oro/’oro<n>$ ^levu/levu<adj>$ ^yai/yai<det><dem>$^./.<sent>$</code>
:-English(desired):<code>^this/this<det><dem><sg>$ ^big/big<adj><sint>$ ^village/village<n><sg>$^./.<sent>$</code>
*Biltrans:<code>^a<art>/$ ^’oro<n>/village<n>$ ^levu<adj>/big<adj><sint>$ ^yai<det><dem>/this<det><dem>$^.<sent>/.<sent>$</code>
Fijian nouns and demonstratives do not distinguish numbers, so I want to add a number tag for the English noun and the demonstrative ''this'', and then change the word order from head first in Fijian to head final in English:
<pre>
<rule comment="REGLA: NPs with demonstratives">
<pattern>
<pattern-item n="article"/>
<pattern-item n="noun_reg"/>
<pattern-item n="adj"/>
<pattern-item n="dem"/>
</pattern>
<action>
<out>
<chunk name="nom_adj_dem" case="caseFirstWord">
<tags>
<tag><lit-tag v="SN"/></tag>
<tag><lit-tag v="sg"/></tag>
</tags>
<lu>
<clip pos="4" side="tl" part="lem"/>
<clip pos="4" side="tl" part="a_pos"/>
<clip pos="4" side="tl" part="subcategories"/>
<lit-tag v="sg"/>
</lu>

<lu>
<clip pos="3" side="tl" part="whole"/>
</lu>

<lu>
<clip pos="2" side="tl" part="lem"/>
<clip pos="2" side="tl" part="a_pos"/>
<lit-tag v="sg"/>
</lu>
</chunk>
</out>
</action>
</rule>
</pre>
*Chunker:<code>^nom_adj_dem<SN><sg>{^this<det><dem><sg>$ ^big<adj><sint>$ ^village<n><sg>$}$^sent<SENT>{^.<sent>$}$</code>
*Interchunk:<code>^nom_adj_dem<SN><sg>{^this<det><dem><sg>$ ^big<adj><sint>$ ^village<n><sg>$}$^sent<SENT>{^.<sent>$}$</code>
*Postchunk:<code>^this<det><dem><sg>$ ^big<adj><sint>$ ^village<n><sg>$^.<sent>$</code>
*Note and problems with current codes:
Before this I tried to add number tags by having <code><clip pos="2" side="tl" part="a_num"></code> in the lexical unit section, but since Fijian nouns do not change for numbers (e.g. ’oro can be translated as both 'village' and 'villages'), the English noun still did not what number tag to take and thus the number tag never showed up in the chunk. Instead, I added the number tag manually to the noun and demonstrative in the English translation by <lit-tag>, setting it to be "sg" in this pattern. It seems to work in this case, but in fact without contexts from sentences, it is inaccurate to translate "a ’oro levu yai" as "this big village" (e.g. {{transferTest|fij|eng|Au raica (koya) a ’oro levu yai|I see this big village.}} and {{transferTest|fij|eng|Au raica ira a ’oro levu yai|I see these big villages.}}). So I need to change my example from a phrase to a sentence with enough contexts and write a rule for that pattern instead.

==Post-evaluation==
*eng → fij
:-WER: 66.13%
:-PER: 61.29%
*fij → eng
:-WER: 75.00%
:-PER: 75.00%

Fijian and English/Structural transfer

2018-04-15T17:53:08Z

Hwang11: /* fij → eng */

==Initial Evaluation==
*eng → fij
:-WER:75.81%
:-PER:70.97%
*fij → eng
:-WER:82.14%
:-PER:76.79%

==eng → fij==
I am working on the first contrastive grammar point--word orders within NPs. An English NP has the demonstrative and adjectives preceding the head noun, while Fijian has all of the modifiers following the head noun. Besides, the article in front of the head noun is obligatory even if a demonstrative is present, but its corresponding translation in English--the determiner ''the''--is not allowed to occur when there is a demonstrative in the NP. In terms of tags, Fijian does not have number tags on either nouns or demonstratives.
*Example phrase: {{transferTest|eng|fij|this big village|a ’oro levu yai}}
*Current translation to Fijian:<code>#yai levu #’oro</code>
*Taggers: <code>^this/this<det><dem><sg>$ ^big/big<adj><sint>$ ^village/village<n><sg>$^./.<sent>$</code>
*Biltrans: <code>^this<det><dem><sg>/yai<det><dem><sg>$ ^big<adj><sint>/levu<adj>$ ^village<n><sg>/’oro<n><sg>$^.<sent>/.<sent>$</code>
Step 1: remove number tags and add ''a<art>'' in the nom chunk:
*Chunker: <code>^dem<det><dem>{^yai<det><dem>$}$ ^adj<adj><sint>{^levu<adj>$}$ ^nom<n>{^a<art> ^’oro<n>$}$^sent<SENT>{^.<sent>$}$</code>
Step 2: change word orders:
*Interchunk: <code>^nom<n>{^a<art> ^’oro<n>$}$ ^adj<adj><sint>{^levu<adj>$}$ ^dem<det><dem>{^yai<det><dem>$}$^sent<SENT>{^.<sent>$}$</code>
*Postchunk: <code>^a<art>$ ^’oro<n>$ ^levu<adj>$ ^yai<det><dem>$^.<sent>$</code>
*Translation: <code>a ’oro levu yai</code>

==fij → eng==
Same grammar point as above.

The relationship between the Fijian article ''a'' and the English definite determiner ''the'' was complicated to implement. For now, I set the article ''a'' to translate to null all the time in the bilingual dictionary. This works here and for most complex NPs such as those with demonstratives and possessives, but not for NPs without any <det>, such as ''a ’oro levu'' (ART village big) or ''a ’oro'' (ART village); for these simple NPs, I will need to write separate rules with the two patterns in the fij-eng.t1x file.
*Example phrase: {{transferTest|fij|eng|a ’oro levu yai|this big village}}
*Current translation from Fijian to English: <code>#village big #this</code>
*Taggers:
:-Fijian:<code>^a/a<art>$ ^’oro/’oro<n>$ ^levu/levu<adj>$ ^yai/yai<det><dem>$^./.<sent>$</code>
:-English(desired):<code>^this/this<det><dem><sg>$ ^big/big<adj><sint>$ ^village/village<n><sg>$^./.<sent>$</code>
*Biltrans:<code>^a<art>/$ ^’oro<n>/village<n>$ ^levu<adj>/big<adj><sint>$ ^yai<det><dem>/this<det><dem>$^.<sent>/.<sent>$</code>
Fijian nouns and demonstratives do not distinguish numbers, so I want to add a number tag for the English noun and the demonstrative ''this'', and then change the word order from head first in Fijian to head final in English:
<pre>
<rule comment="REGLA: NPs with demonstratives">
<pattern>
<pattern-item n="article"/>
<pattern-item n="noun_reg"/>
<pattern-item n="adj"/>
<pattern-item n="dem"/>
</pattern>
<action>
<out>
<chunk name="nom_adj_dem" case="caseFirstWord">
<tags>
<tag><lit-tag v="SN"/></tag>
<tag><lit-tag v="sg"/></tag>
</tags>
<lu>
<clip pos="4" side="tl" part="lem"/>
<clip pos="4" side="tl" part="a_pos"/>
<clip pos="4" side="tl" part="subcategories"/>
<lit-tag v="sg"/>
</lu>

<lu>
<clip pos="3" side="tl" part="whole"/>
</lu>

<lu>
<clip pos="2" side="tl" part="lem"/>
<clip pos="2" side="tl" part="a_pos"/>
<lit-tag v="sg"/>
</lu>
</chunk>
</out>
</action>
</rule>
</pre>
*Chunker:<code>^nom_adj_dem<SN><sg>{^this<det><dem><sg>$ ^big<adj><sint>$ ^village<n><sg>$}$^sent<SENT>{^.<sent>$}$</code>
*Interchunk:<code>^nom_adj_dem<SN><sg>{^this<det><dem><sg>$ ^big<adj><sint>$ ^village<n><sg>$}$^sent<SENT>{^.<sent>$}$</code>
*Postchunk:<code>^this<det><dem><sg>$ ^big<adj><sint>$ ^village<n><sg>$^.<sent>$</code>
*Note and problems with current codes:
Before this I tried to add number tags by writing <code><clip pos="2" side="tl" part="a_num"></code> in the lexical unit section, but since Fijian nouns do not change for numbers (e.g. ’oro can be translated as both 'village' and 'villages'), the English noun still did not what number tag to take and thus the number tag never showed up in the chunk. Instead, I added the number tag manually to the noun and demonstrative in the English translation by <lit-tag>, setting it to be "sg" in this pattern. It seems to work in this case, but in fact without contexts from sentences the phrase is in, it is inaccurate to translate "a ’oro levu yai" as "this big village" (e.g. {{transferTest|fij|eng|Au raica koya a ’oro levu yai|I see this big village.}} and {{transferTest|fij|eng|Au raica ira a ’oro levu yai|I see these big villages.}}). So I need to change my example from a phrase to a sentence with enough contexts and write a rule for that pattern instead.

==Post-evaluation==
*eng → fij
:-WER: 66.13%
:-PER: 61.29%
*fij → eng
:-WER: 75.00%
:-PER: 75.00%