http://wikis.swarthmore.edu/ling073/api.php?action=feedcontributions&feedformat=atom&user=Eortici1LING073 - User contributions [en]2026-10-25T18:07:25ZUser contributionsMediaWiki 1.27.7http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5330User:Eortici1/Final project2017-05-10T19:21:23Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger [http://universaldependencies.org/ Universal Dependencies] corpus and annotation guidelines for Bikol.<br />
<br />
The <b><u>[https://github.com/eorticio/UD_Bikol repository]</u></b> for this project is modeled closely after the repositories used by Universal Dependencies ([https://github.com/UniversalDependencies/UD_Greek example here]).<br />
<br />
== Introduction to UD ==<br />
<br />
The Universal Dependencies project was created to "provide a universal inventory of categories and guidelines to facilitate consistent annotation of similar constructions across languages". UD uses a set of typed dependency relations between words, where one word is the head of the sentence and all other words are dependent on another word in the sentence. These dependencies are generally universal and are meant to bring out cross-lingual typological parallels among the world's languages.<br />
<br />
<br />
For examples of the relations used in the Bikol treebank, see the [[#UD Annotation Guidelines|Annotation guidelines for Bikol]].<br />
<br />
For examples of annotated Bikol sentences, see [[#Corpus Information|the Bikol corpus]].<br />
<br />
== Corpus Information ==<br />
<br />
The annotated corpus can be found in [https://github.com/eorticio/UD_Bikol/blob/master/bcl-ud-corpus.txt .txt] and .conllu ([https://github.com/eorticio/UD_Bikol/blob/master/bcl-ud-train.conllu with] and [https://github.com/eorticio/UD_Bikol/blob/master/bcl-ud-train-nomorph.conllu without] morphological information) in the project repository.<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Tokens<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
<br />
The corpus contains extracts from the following sources:<br />
<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 08/05/2017<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
<br />
== Process ==<br />
<br />
In the following section, I address solutions to specific problems I encountered while annotating my corpus.<br />
<br />
=== Pluralization of nouns with 'mga' ===<br />
<br />
To make nouns plural in Bikol, you add the word 'mga' before it.<br />
Halion mo an <i>mga papers</i> sa lamesa ta pigkorehiran na iyan.<br />
<br />
When deciding which relation to use for this scenario, my initial thought was to use the [http://universaldependencies.org/u/dep/nummod.html @nummod (numeric modifier) relation], because this is a separate word that modifies the quantity of a noun. However, 'mga' has no isolated meaning in Bikol and is therefore not a numeral. So, I decided to use the [http://universaldependencies.org/u/dep/det.html @det (determiner) relation]. In UD's documentation of what constitutes a determiner, it says that quantity determiners (quantifiers) qualify for this relation.<br />
<br />
=== Linker in Modification Structures: 'na' ===<br />
<br />
The biggest challenge of this project (besides the time constraint) was deciding on the relation to use for the linking word, 'na'. In the grammar books I have, this word is simply described as a linking word in modification structures. The word is used very frequently in Bikol in a variety of different modification structures. For this word, I decided to create four [http://universaldependencies.org/ext-dep-index.html language-specific relations]. These are outlined below. <br />
<br />
==== amod ====<br />
May <b>turista na puti</b> nagduman sa San Jose.<br />
<i>There is a white tourist that visited San Jose.</i><br />
<br />
This sentence demonstrates an @amod relation, in which an adjectival modifier (puti, or white) modifies the meaning of a nominal (turista, or tourist). The linking word 'na' is assigned an aux:amod relation dependent on the modifier. Language-specific relations are, by convention, composed in this manner (universal:extension). I chose to make this relation an extension of the @aux (auxiliary) relation because the linking word serves as an auxiliary word to the modifier.<br />
<br />
For more documentation on this relation, see the [https://github.com/eorticio/UD_Bikol/blob/master/bcl-dep/aux-amod.md aux:amod documentation].<br />
<br />
==== nmod ====<br />
<b>kapitolyo na ekonomiya</b> nin Albay<br />
<i>economic capital of Albay</i><br />
<br />
This sentence demonstrates an @nmod relation, in which a nominal modifier (ekonomiya) acts upon another nominal (kapitolyo). The linking word 'na' is assigned an aux:nmod relation dependent on the modifier.<br />
<br />
For more documentation on this relation, see the [https://github.com/eorticio/UD_Bikol/blob/master/bcl-dep/aux-nmod.md aux:nmod documentation].<br />
<br />
==== advmod ====<br />
<b>Matuga na naman</b> siguro an bulkan Mayon.<br />
<i>The Mayon volcano will surely erupt again.</i><br />
<br />
This sentence demonstrates an @advmod relation, in which an adverbial modifier (naman, or again) acts upon the predicate, matuga (will erupt). The linking word 'na' is assigned an aux:advmod relation dependent on the adverbial modifier.<br />
<br />
Note that siguro (surely), is also an adverbial modifier and also has an @advmod relation to matuga. In this case of multiple adverbial modifiers, only one linking word is necessary. Therefore, the linking word is given the aux:advmod relation to the first adverb because that is the one it is physically linking to the predicate.<br />
<br />
For more documentation on this relation, see the [https://github.com/eorticio/UD_Bikol/blob/master/bcl-dep/aux-advmod.md aux:advmod documentation].<br />
<br />
==== nummod ====<br />
Ini may <b>apat na eskalon</b>.<br />
<i>This has four floors.</i><br />
<br />
This sentence demonstrates a @nummod relation, in which an numeric modifier (apat, four) acts upon a nominal (eskalon, or floor). The linking word 'na' is assigned an aux:nummod relation dependent on the adjectival modifier.<br />
<br />
Note the order here. Typically, modifiers come after the thing that they modify. However, for @nummod relations, it is typically the opposite, with the numeric modifier coming first. The linking word thus comes after the modifier in order to remain in between the modifier and the word it modifies.<br />
<br />
For more documentation on this relation, see the [https://github.com/eorticio/UD_Bikol/blob/master/bcl-dep/aux-nummod.md aux:nummod documentation].<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
The final component of the project was to create annotation guidelines for the Bikol UD Treebank. This documentation would be useful to anyone who wished to contribute to the Treebank.<br />
<br />
Again, I used the same format that is used by the Universal Dependencies Project, using the [https://github.com/UniversalDependencies/docs/tree/pages-source/_sv-dep repository of Swedish guidelines] as a guide.<br />
<br />
The annotation guidelines can be found in the [https://github.com/eorticio/UD_Bikol/tree/master/bcl-dep bcl-dep directory] of the project's repository, and feature documentation on 25 of the most commonly used relations I used in my corpus, including the four language-specific relations I mentioned earlier in the [[#Process|Process section]].<br />
<br />
== Evaluation ==<br />
<br />
Parsers are typically evaluated with the LAS (Label Attachment Score) and UAS (Unlabeled Attachment Score) obtained by testing on a separate corpus. However, making an additional, separate corpus would be time consuming. To get around this, I divided the corpus into 10 roughly equal parts (10-11 sentences each). Then, I used each part as a test corpus for two parsers (with and without morphological information) which were trained on the remaining corpus. <br />
<br />
The partial corpora, conllu and udpipe files can all be found in the [https://github.com/eorticio/UD_Bikol/tree/master/bcl-eval bcl-eval directory] of the repository. The data from all the individual tests can be found in the [[#Appendix|Appendix]].<br />
<br />
To evaluate the [https://github.com/eorticio/UD_Bikol/tree/master/dev parsers trained on the entire corpus], I took the average of all the UAS/LAS scores obtained on the 10 test sets, ± the standard deviation. The results are displayed below.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-withmorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|60.87 ± 5.67<br />
|50.00 ± 4.22<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-nomorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|50.51 ± 5.53<br />
|36.01 ± 6.59<br />
|}<br />
<br />
== Further Work ==<br />
<br />
Further work on this project would mainly involve increasing the size of the annotated corpus. My corpus of ~100 sentences and ~1200 tokens is very small in comparison to completed UD Treebanks, due to the time constraints of the project and the nature of the material I was working with (Wikipedia articles, no glosses). The most fully developed UD treebanks have 100-200K+ tokens annotated, which would yield much, much more accurate parsers.<br />
<br />
== Appendix ==<br />
<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation: Tests on 10 Subsets of Corpus<br />
|-<br />
! <br />
! UAS with morphology<br />
! LAS with morphology<br />
! UAS without morphology<br />
! LAS without morphology<br />
|-<br />
! Test 1<br />
| 55.37<br />
| 45.45<br />
| 48.76<br />
| 31.40<br />
|-<br />
! Test 2<br />
| 56.34<br />
| 46.48<br />
| 50.70<br />
| 38.03<br />
|-<br />
! Test 3<br />
| 64.29<br />
| 54.46<br />
| 57.14<br />
| 43.75<br />
|-<br />
! Test 4<br />
| 65.00<br />
| 49.17<br />
| 53.33<br />
| 36.67<br />
|-<br />
! Test 5<br />
| 56.44<br />
| 47.52<br />
| 40.59<br />
| 23.76<br />
|-<br />
! Test 6<br />
| 65.52<br />
| 52.41<br />
| 56.55<br />
| 38.62<br />
|-<br />
! Test 7<br />
| 52.89<br />
| 44.63<br />
| 46.28<br />
| 31.40<br />
|-<br />
! Test 8<br />
| 57.45<br />
| 48.94<br />
| 45.74<br />
| 34.04<br />
|-<br />
! Test 9<br />
| 67.65<br />
| 53.92<br />
| 49.02<br />
| 35.29<br />
|-<br />
! Test 10<br />
| 67.77<br />
| 57.02<br />
| 57.02<br />
| 47.11<br />
|}<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5329User:Eortici1/Final project2017-05-10T17:39:26Z<p>Eortici1: /* nummod */</p>
<hr />
<div>This page documents my final project, a larger [http://universaldependencies.org/ Universal Dependencies] corpus and annotation guidelines for Bikol.<br />
<br />
The <b><u>[https://github.swarthmore.edu/eortici1/UD_Bikol repository]</u></b> for this project is modeled closely after the repositories used by Universal Dependencies ([https://github.com/UniversalDependencies/UD_Greek example here]).<br />
<br />
== Introduction to UD ==<br />
<br />
The Universal Dependencies project was created to "provide a universal inventory of categories and guidelines to facilitate consistent annotation of similar constructions across languages". UD uses a set of typed dependency relations between words, where one word is the head of the sentence and all other words are dependent on another word in the sentence. These dependencies are generally universal and are meant to bring out cross-lingual typological parallels among the world's languages.<br />
<br />
<br />
For examples of the relations used in the Bikol treebank, see the [[#UD Annotation Guidelines|Annotation guidelines for Bikol]].<br />
<br />
For examples of annotated Bikol sentences, see [[#Corpus Information|the Bikol corpus]].<br />
<br />
== Corpus Information ==<br />
<br />
The annotated corpus can be found in [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-corpus.txt .txt] and .conllu ([https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-train.conllu with] and [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-train-nomorph.conllu without] morphological information) in the project repository.<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Tokens<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
<br />
The corpus contains extracts from the following sources:<br />
<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 08/05/2017<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
<br />
== Process ==<br />
<br />
In the following section, I address solutions to specific problems I encountered while annotating my corpus.<br />
<br />
=== Pluralization of nouns with 'mga' ===<br />
<br />
To make nouns plural in Bikol, you add the word 'mga' before it.<br />
Halion mo an <i>mga papers</i> sa lamesa ta pigkorehiran na iyan.<br />
<br />
When deciding which relation to use for this scenario, my initial thought was to use the [http://universaldependencies.org/u/dep/nummod.html @nummod (numeric modifier) relation], because this is a separate word that modifies the quantity of a noun. However, 'mga' has no isolated meaning in Bikol and is therefore not a numeral. So, I decided to use the [http://universaldependencies.org/u/dep/det.html @det (determiner) relation]. In UD's documentation of what constitutes a determiner, it says that quantity determiners (quantifiers) qualify for this relation.<br />
<br />
=== Linker in Modification Structures: 'na' ===<br />
<br />
The biggest challenge of this project (besides the time constraint) was deciding on the relation to use for the linking word, 'na'. In the grammar books I have, this word is simply described as a linking word in modification structures. The word is used very frequently in Bikol in a variety of different modification structures. For this word, I decided to create four [http://universaldependencies.org/ext-dep-index.html language-specific relations]. These are outlined below. <br />
<br />
==== amod ====<br />
May <b>turista na puti</b> nagduman sa San Jose.<br />
<i>There is a white tourist that visited San Jose.</i><br />
<br />
This sentence demonstrates an @amod relation, in which an adjectival modifier (puti, or white) modifies the meaning of a nominal (turista, or tourist). The linking word 'na' is assigned an aux:amod relation dependent on the modifier. Language-specific relations are, by convention, composed in this manner (universal:extension). I chose to make this relation an extension of the @aux (auxiliary) relation because the linking word serves as an auxiliary word to the modifier.<br />
<br />
For more documentation on this relation, see the [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-dep/aux-amod.md aux:amod documentation].<br />
<br />
==== nmod ====<br />
<b>kapitolyo na ekonomiya</b> nin Albay<br />
<i>economic capital of Albay</i><br />
<br />
This sentence demonstrates an @nmod relation, in which a nominal modifier (ekonomiya) acts upon another nominal (kapitolyo). The linking word 'na' is assigned an aux:nmod relation dependent on the modifier.<br />
<br />
For more documentation on this relation, see the [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-dep/aux-nmod.md aux:nmod documentation].<br />
<br />
==== advmod ====<br />
<b>Matuga na naman</b> siguro an bulkan Mayon.<br />
<i>The Mayon volcano will surely erupt again.</i><br />
<br />
This sentence demonstrates an @advmod relation, in which an adverbial modifier (naman, or again) acts upon the predicate, matuga (will erupt). The linking word 'na' is assigned an aux:advmod relation dependent on the adverbial modifier.<br />
<br />
Note that siguro (surely), is also an adverbial modifier and also has an @advmod relation to matuga. In this case of multiple adverbial modifiers, only one linking word is necessary. Therefore, the linking word is given the aux:advmod relation to the first adverb because that is the one it is physically linking to the predicate.<br />
<br />
For more documentation on this relation, see the [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-dep/aux-advmod.md aux:advmod documentation].<br />
<br />
==== nummod ====<br />
Ini may <b>apat na eskalon</b>.<br />
<i>This has four floors.</i><br />
<br />
This sentence demonstrates a @nummod relation, in which an numeric modifier (apat, four) acts upon a nominal (eskalon, or floor). The linking word 'na' is assigned an aux:nummod relation dependent on the adjectival modifier.<br />
<br />
Note the order here. Typically, modifiers come after the thing that they modify. However, for @nummod relations, it is typically the opposite, with the numeric modifier coming first. The linking word thus comes after the modifier in order to remain in between the modifier and the word it modifies.<br />
<br />
For more documentation on this relation, see the [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-dep/aux-nummod.md aux:nummod documentation].<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
The final component of the project was to create annotation guidelines for the Bikol UD Treebank. This documentation would be useful to anyone who wished to contribute to the Treebank.<br />
<br />
Again, I used the same format that is used by the Universal Dependencies Project, using the [https://github.com/UniversalDependencies/docs/tree/pages-source/_sv-dep repository of Swedish guidelines] as a guide.<br />
<br />
The annotation guidelines can be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-dep bcl-dep directory] of the project's repository, and feature documentation on 25 of the most commonly used relations I used in my corpus, including the four language-specific relations I mentioned earlier in the [[#Process|Process section]].<br />
<br />
== Evaluation ==<br />
<br />
Parsers are typically evaluated with the LAS (Label Attachment Score) and UAS (Unlabeled Attachment Score) obtained by testing on a separate corpus. However, making an additional, separate corpus would be time consuming. To get around this, I divided the corpus into 10 roughly equal parts (10-11 sentences each). Then, I used each part as a test corpus for two parsers (with and without morphological information) which were trained on the remaining corpus. <br />
<br />
The partial corpora, conllu and udpipe files can all be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-eval bcl-eval directory] of the repository. The data from all the individual tests can be found in the [[#Appendix|Appendix]].<br />
<br />
To evaluate the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/dev parsers trained on the entire corpus], I took the average of all the UAS/LAS scores obtained on the 10 test sets, ± the standard deviation. The results are displayed below.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-withmorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|60.87 ± 5.67<br />
|50.00 ± 4.22<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-nomorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|50.51 ± 5.53<br />
|36.01 ± 6.59<br />
|}<br />
<br />
== Further Work ==<br />
<br />
Further work on this project would mainly involve increasing the size of the annotated corpus. My corpus of ~100 sentences and ~1200 tokens is very small in comparison to completed UD Treebanks, due to the time constraints of the project and the nature of the material I was working with (Wikipedia articles, no glosses). The most fully developed UD treebanks have 100-200K+ tokens annotated, which would yield much, much more accurate parsers.<br />
<br />
== Appendix ==<br />
<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation: Tests on 10 Subsets of Corpus<br />
|-<br />
! <br />
! UAS with morphology<br />
! LAS with morphology<br />
! UAS without morphology<br />
! LAS without morphology<br />
|-<br />
! Test 1<br />
| 55.37<br />
| 45.45<br />
| 48.76<br />
| 31.40<br />
|-<br />
! Test 2<br />
| 56.34<br />
| 46.48<br />
| 50.70<br />
| 38.03<br />
|-<br />
! Test 3<br />
| 64.29<br />
| 54.46<br />
| 57.14<br />
| 43.75<br />
|-<br />
! Test 4<br />
| 65.00<br />
| 49.17<br />
| 53.33<br />
| 36.67<br />
|-<br />
! Test 5<br />
| 56.44<br />
| 47.52<br />
| 40.59<br />
| 23.76<br />
|-<br />
! Test 6<br />
| 65.52<br />
| 52.41<br />
| 56.55<br />
| 38.62<br />
|-<br />
! Test 7<br />
| 52.89<br />
| 44.63<br />
| 46.28<br />
| 31.40<br />
|-<br />
! Test 8<br />
| 57.45<br />
| 48.94<br />
| 45.74<br />
| 34.04<br />
|-<br />
! Test 9<br />
| 67.65<br />
| 53.92<br />
| 49.02<br />
| 35.29<br />
|-<br />
! Test 10<br />
| 67.77<br />
| 57.02<br />
| 57.02<br />
| 47.11<br />
|}<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5328User:Eortici1/Final project2017-05-10T17:39:07Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger [http://universaldependencies.org/ Universal Dependencies] corpus and annotation guidelines for Bikol.<br />
<br />
The <b><u>[https://github.swarthmore.edu/eortici1/UD_Bikol repository]</u></b> for this project is modeled closely after the repositories used by Universal Dependencies ([https://github.com/UniversalDependencies/UD_Greek example here]).<br />
<br />
== Introduction to UD ==<br />
<br />
The Universal Dependencies project was created to "provide a universal inventory of categories and guidelines to facilitate consistent annotation of similar constructions across languages". UD uses a set of typed dependency relations between words, where one word is the head of the sentence and all other words are dependent on another word in the sentence. These dependencies are generally universal and are meant to bring out cross-lingual typological parallels among the world's languages.<br />
<br />
<br />
For examples of the relations used in the Bikol treebank, see the [[#UD Annotation Guidelines|Annotation guidelines for Bikol]].<br />
<br />
For examples of annotated Bikol sentences, see [[#Corpus Information|the Bikol corpus]].<br />
<br />
== Corpus Information ==<br />
<br />
The annotated corpus can be found in [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-corpus.txt .txt] and .conllu ([https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-train.conllu with] and [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-train-nomorph.conllu without] morphological information) in the project repository.<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Tokens<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
<br />
The corpus contains extracts from the following sources:<br />
<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 08/05/2017<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
<br />
== Process ==<br />
<br />
In the following section, I address solutions to specific problems I encountered while annotating my corpus.<br />
<br />
=== Pluralization of nouns with 'mga' ===<br />
<br />
To make nouns plural in Bikol, you add the word 'mga' before it.<br />
Halion mo an <i>mga papers</i> sa lamesa ta pigkorehiran na iyan.<br />
<br />
When deciding which relation to use for this scenario, my initial thought was to use the [http://universaldependencies.org/u/dep/nummod.html @nummod (numeric modifier) relation], because this is a separate word that modifies the quantity of a noun. However, 'mga' has no isolated meaning in Bikol and is therefore not a numeral. So, I decided to use the [http://universaldependencies.org/u/dep/det.html @det (determiner) relation]. In UD's documentation of what constitutes a determiner, it says that quantity determiners (quantifiers) qualify for this relation.<br />
<br />
=== Linker in Modification Structures: 'na' ===<br />
<br />
The biggest challenge of this project (besides the time constraint) was deciding on the relation to use for the linking word, 'na'. In the grammar books I have, this word is simply described as a linking word in modification structures. The word is used very frequently in Bikol in a variety of different modification structures. For this word, I decided to create four [http://universaldependencies.org/ext-dep-index.html language-specific relations]. These are outlined below. <br />
<br />
==== amod ====<br />
May <b>turista na puti</b> nagduman sa San Jose.<br />
<i>There is a white tourist that visited San Jose.</i><br />
<br />
This sentence demonstrates an @amod relation, in which an adjectival modifier (puti, or white) modifies the meaning of a nominal (turista, or tourist). The linking word 'na' is assigned an aux:amod relation dependent on the modifier. Language-specific relations are, by convention, composed in this manner (universal:extension). I chose to make this relation an extension of the @aux (auxiliary) relation because the linking word serves as an auxiliary word to the modifier.<br />
<br />
For more documentation on this relation, see the [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-dep/aux-amod.md aux:amod documentation].<br />
<br />
==== nmod ====<br />
<b>kapitolyo na ekonomiya</b> nin Albay<br />
<i>economic capital of Albay</i><br />
<br />
This sentence demonstrates an @nmod relation, in which a nominal modifier (ekonomiya) acts upon another nominal (kapitolyo). The linking word 'na' is assigned an aux:nmod relation dependent on the modifier.<br />
<br />
For more documentation on this relation, see the [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-dep/aux-nmod.md aux:nmod documentation].<br />
<br />
==== advmod ====<br />
<b>Matuga na naman</b> siguro an bulkan Mayon.<br />
<i>The Mayon volcano will surely erupt again.</i><br />
<br />
This sentence demonstrates an @advmod relation, in which an adverbial modifier (naman, or again) acts upon the predicate, matuga (will erupt). The linking word 'na' is assigned an aux:advmod relation dependent on the adverbial modifier.<br />
<br />
Note that siguro (surely), is also an adverbial modifier and also has an @advmod relation to matuga. In this case of multiple adverbial modifiers, only one linking word is necessary. Therefore, the linking word is given the aux:advmod relation to the first adverb because that is the one it is physically linking to the predicate.<br />
<br />
For more documentation on this relation, see the [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-dep/aux-advmod.md aux:advmod documentation].<br />
<br />
==== nummod ====<br />
Ini may <b>apat na eskalon</b>.<br />
<i>This has four floors.</i><br />
<br />
This sentence demonstrates a @nummod relation, in which an numeric modifier (apat, four) acts upon a nominal (eskalon, or floor). The linking word 'na' is assigned an aux:nummod relation dependent on the adjectival modifier.<br />
<br />
Note the order here. Typically, modifiers come after the thing that they modify. However, for @nummod relations, it is typically the opposite, with the numeric modifier coming first. The linking word thus comes after the modifier in order to remain in between the modifier and the word it modifies.<br />
<br />
For more documentation on this relation, see the [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-dep/aux-nummod.md aux:nummod documentation].<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
The final component of the project was to create annotation guidelines for the Bikol UD Treebank. This documentation would be useful to anyone who wished to contribute to the Treebank.<br />
<br />
Again, I used the same format that is used by the Universal Dependencies Project, using the [https://github.com/UniversalDependencies/docs/tree/pages-source/_sv-dep repository of Swedish guidelines] as a guide.<br />
<br />
The annotation guidelines can be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-dep bcl-dep directory] of the project's repository, and feature documentation on 25 of the most commonly used relations I used in my corpus, including the four language-specific relations I mentioned earlier in the [[#Process|Process section]].<br />
<br />
== Evaluation ==<br />
<br />
Parsers are typically evaluated with the LAS (Label Attachment Score) and UAS (Unlabeled Attachment Score) obtained by testing on a separate corpus. However, making an additional, separate corpus would be time consuming. To get around this, I divided the corpus into 10 roughly equal parts (10-11 sentences each). Then, I used each part as a test corpus for two parsers (with and without morphological information) which were trained on the remaining corpus. <br />
<br />
The partial corpora, conllu and udpipe files can all be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-eval bcl-eval directory] of the repository. The data from all the individual tests can be found in the [[#Appendix|Appendix]].<br />
<br />
To evaluate the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/dev parsers trained on the entire corpus], I took the average of all the UAS/LAS scores obtained on the 10 test sets, ± the standard deviation. The results are displayed below.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-withmorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|60.87 ± 5.67<br />
|50.00 ± 4.22<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-nomorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|50.51 ± 5.53<br />
|36.01 ± 6.59<br />
|}<br />
<br />
== Further Work ==<br />
<br />
Further work on this project would mainly involve increasing the size of the annotated corpus. My corpus of ~100 sentences and ~1200 tokens is very small in comparison to completed UD Treebanks, due to the time constraints of the project and the nature of the material I was working with (Wikipedia articles, no glosses). The most fully developed UD treebanks have 100-200K+ tokens annotated, which would yield much, much more accurate parsers.<br />
<br />
== Appendix ==<br />
<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation: Tests on 10 Subsets of Corpus<br />
|-<br />
! <br />
! UAS with morphology<br />
! LAS with morphology<br />
! UAS without morphology<br />
! LAS without morphology<br />
|-<br />
! Test 1<br />
| 55.37<br />
| 45.45<br />
| 48.76<br />
| 31.40<br />
|-<br />
! Test 2<br />
| 56.34<br />
| 46.48<br />
| 50.70<br />
| 38.03<br />
|-<br />
! Test 3<br />
| 64.29<br />
| 54.46<br />
| 57.14<br />
| 43.75<br />
|-<br />
! Test 4<br />
| 65.00<br />
| 49.17<br />
| 53.33<br />
| 36.67<br />
|-<br />
! Test 5<br />
| 56.44<br />
| 47.52<br />
| 40.59<br />
| 23.76<br />
|-<br />
! Test 6<br />
| 65.52<br />
| 52.41<br />
| 56.55<br />
| 38.62<br />
|-<br />
! Test 7<br />
| 52.89<br />
| 44.63<br />
| 46.28<br />
| 31.40<br />
|-<br />
! Test 8<br />
| 57.45<br />
| 48.94<br />
| 45.74<br />
| 34.04<br />
|-<br />
! Test 9<br />
| 67.65<br />
| 53.92<br />
| 49.02<br />
| 35.29<br />
|-<br />
! Test 10<br />
| 67.77<br />
| 57.02<br />
| 57.02<br />
| 47.11<br />
|}<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5294User:Eortici1/Final project2017-05-10T04:52:43Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger [http://universaldependencies.org/ Universal Dependencies] corpus and annotation guidelines for Bikol.<br />
<br />
The <b><u>[https://github.swarthmore.edu/eortici1/UD_Bikol repository]</u></b> for this project is modeled closely after the repositories used by Universal Dependencies ([https://github.com/UniversalDependencies/UD_Greek example here]).<br />
<br />
== Introduction to UD ==<br />
<br />
The Universal Dependencies project was created to "provide a universal inventory of categories and guidelines to facilitate consistent annotation of similar constructions across languages". UD uses a set of typed dependency relations between words, where one word is the head of the sentence and all other words are dependent on another word in the sentence. These dependencies are generally universal and are meant to bring out cross-lingual typological parallels among the world's languages.<br />
<br />
<br />
For examples of the relations used in the Bikol treebank, see the [[#UD Annotation Guidelines|Annotation guidelines for Bikol]].<br />
<br />
For examples of annotated Bikol sentences, see [[#Corpus Information|the Bikol corpus]].<br />
<br />
== Corpus Information ==<br />
<br />
The annotated corpus can be found in [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-corpus.txt .txt] and .conllu ([https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-train.conllu with] and [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-train-nomorph.conllu without] morphological information) in the project repository.<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Tokens<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
<br />
The corpus contains extracts from the following sources:<br />
<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 08/05/2017<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
<br />
== Process ==<br />
<br />
In the following section, I address solutions to specific problems I encountered while annotating my corpus.<br />
<br />
=== Pluralization of nouns with 'mga' ===<br />
<br />
To make nouns plural in Bikol, you add the word 'mga' before it.<br />
Halion mo an <i>mga papers</i> sa lamesa ta pigkorehiran na iyan.<br />
<br />
When deciding which relation to use for this scenario, my initial thought was to use the [http://universaldependencies.org/u/dep/nummod.html @nummod (numeric modifier) relation], because this is a separate word that modifies the quantity of a noun. However, 'mga' has no isolated meaning in Bikol and is therefore not a numeral. So, I decided to use the [http://universaldependencies.org/u/dep/det.html @det (determiner) relation]. In UD's documentation of what constitutes a determiner, it says that quantity determiners (quantifiers) qualify for this relation.<br />
<br />
=== Linker in Modification Structures: 'na' ===<br />
<br />
The biggest challenge of this project (besides the time constraint) was deciding on the relation to use for the linking word, 'na'. In the grammar books I have, this word is simply described as a linking word in modification structures. The word is used very frequently in Bikol in a variety of different modification structures. For this word, I decided to create four [http://universaldependencies.org/ext-dep-index.html language-specific relations]. These are outlined below. <br />
<br />
==== amod ====<br />
May <b>turista na puti</b> nagduman sa San Jose.<br />
<i>There is a white tourist that visited San Jose.</i><br />
<br />
This sentence demonstrates an @amod relation, in which an adjectival modifier (puti, or white) modifies the meaning of a nominal (turista, or tourist). The linking word 'na' is assigned an aux:amod relation dependent on the modifier. Language-specific relations are, by convention, composed in this manner (universal:extension). I chose to make this relation an extension of the @aux (auxiliary) relation because the linking word serves as an auxiliary word to the modifier.<br />
<br />
For more documentation on this relation, see the aux:amod documentation.<br />
<br />
==== nmod ====<br />
<br />
==== advmod ====<br />
<br />
==== nummod ====<br />
<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
The final component of the project was to create annotation guidelines for the Bikol UD Treebank. This documentation would be useful to anyone who wished to contribute to the Treebank.<br />
<br />
Again, I used the same format that is used by the Universal Dependencies Project, using the [https://github.com/UniversalDependencies/docs/tree/pages-source/_sv-dep repository of Swedish guidelines] as a guide.<br />
<br />
The annotation guidelines can be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-dep bcl-dep directory] of the project's repository, and feature documentation on 25 of the most commonly used relations I used in my corpus, including the four language-specific relations I mentioned earlier in the [[#Process|Process section]].<br />
<br />
== Evaluation ==<br />
<br />
Parsers are typically evaluated with the LAS (Label Attachment Score) and UAS (Unlabeled Attachment Score) obtained by testing on a separate corpus. However, making an additional, separate corpus would be time consuming. To get around this, I divided the corpus into 10 roughly equal parts (10-11 sentences each). Then, I used each part as a test corpus for two parsers (with and without morphological information) which were trained on the remaining corpus. <br />
<br />
The partial corpora, conllu and udpipe files can all be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-eval bcl-eval directory] of the repository. The data from all the individual tests can be found in the [[#Appendix|Appendix]].<br />
<br />
To evaluate the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/dev parsers trained on the entire corpus], I took the average of all the UAS/LAS scores obtained on the 10 test sets, ± the standard deviation. The results are displayed below.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-withmorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|60.87 ± 5.67<br />
|50.00 ± 4.22<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-nomorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|50.51 ± 5.53<br />
|36.01 ± 6.59<br />
|}<br />
<br />
== Further Work ==<br />
<br />
Further work on this project would mainly involve increasing the size of the annotated corpus. My corpus of ~100 sentences and ~1200 tokens is very small in comparison to completed UD Treebanks, due to the time constraints of the project and the nature of the material I was working with (Wikipedia articles, no glosses). The most fully developed UD treebanks have 100-200K+ tokens annotated, which would yield much, much more accurate parsers.<br />
<br />
== Appendix ==<br />
<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation: Tests on 10 Subsets of Corpus<br />
|-<br />
! <br />
! UAS with morphology<br />
! LAS with morphology<br />
! UAS without morphology<br />
! LAS without morphology<br />
|-<br />
! Test 1<br />
| 55.37<br />
| 45.45<br />
| 48.76<br />
| 31.40<br />
|-<br />
! Test 2<br />
| 56.34<br />
| 46.48<br />
| 50.70<br />
| 38.03<br />
|-<br />
! Test 3<br />
| 64.29<br />
| 54.46<br />
| 57.14<br />
| 43.75<br />
|-<br />
! Test 4<br />
| 65.00<br />
| 49.17<br />
| 53.33<br />
| 36.67<br />
|-<br />
! Test 5<br />
| 56.44<br />
| 47.52<br />
| 40.59<br />
| 23.76<br />
|-<br />
! Test 6<br />
| 65.52<br />
| 52.41<br />
| 56.55<br />
| 38.62<br />
|-<br />
! Test 7<br />
| 52.89<br />
| 44.63<br />
| 46.28<br />
| 31.40<br />
|-<br />
! Test 8<br />
| 57.45<br />
| 48.94<br />
| 45.74<br />
| 34.04<br />
|-<br />
! Test 9<br />
| 67.65<br />
| 53.92<br />
| 49.02<br />
| 35.29<br />
|-<br />
! Test 10<br />
| 67.77<br />
| 57.02<br />
| 57.02<br />
| 47.11<br />
|}<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5291User:Eortici1/Final project2017-05-10T04:35:57Z<p>Eortici1: /* Introduction to UD */</p>
<hr />
<div>This page documents my final project, a larger [http://universaldependencies.org/ Universal Dependencies] corpus and annotation guidelines for Bikol.<br />
<br />
The <b><u>[https://github.swarthmore.edu/eortici1/UD_Bikol repository]</u></b> for this project is modeled closely after the repositories used by Universal Dependencies ([https://github.com/UniversalDependencies/UD_Greek example here]).<br />
<br />
== Introduction to UD ==<br />
<br />
The Universal Dependencies project was created to "provide a universal inventory of categories and guidelines to facilitate consistent annotation of similar constructions across languages". UD uses a set of typed dependency relations between words, where one word is the head of the sentence and all other words are dependent on another word in the sentence. These dependencies are generally universal and are meant to bring out cross-lingual typological parallels among the world's languages.<br />
<br />
<br />
For examples of the relations used in the Bikol treebank, see the [[#UD Annotation Guidelines|Annotation guidelines for Bikol]].<br />
<br />
For examples of annotated Bikol sentences, see [[#Corpus Information|the Bikol corpus]].<br />
<br />
== Corpus Information ==<br />
<br />
The annotated corpus can be found in [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-corpus.txt .txt] and .conllu ([https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-train.conllu with] and [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-train-nomorph.conllu without] morphological information) in the project repository.<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Tokens<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
<br />
The corpus contains extracts from the following sources:<br />
<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 08/05/2017<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
<br />
== Process ==<br />
<br />
In the following section, I address solutions to specific problems I encountered while annotating my corpus.<br />
<br />
=== Pluralization of nouns with 'mga' ===<br />
<br />
=== Linker in Modification Structures: 'na' ===<br />
<br />
<br />
* aux:amod, aux:nmod, aux:advmod, aux:nummod<br />
* check @dep and anything labeled !*<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
The final component of the project was to create annotation guidelines for the Bikol UD Treebank. This documentation would be useful to anyone who wished to contribute to the Treebank.<br />
<br />
Again, I used the same format that is used by the Universal Dependencies Project, using the [https://github.com/UniversalDependencies/docs/tree/pages-source/_sv-dep repository of Swedish guidelines] as a guide.<br />
<br />
The annotation guidelines can be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-dep bcl-dep directory] of the project's repository, and feature documentation on 25 of the most commonly used relations I used in my corpus, including the four language-specific relations I mentioned earlier in the [[#Process|Process section]].<br />
<br />
== Evaluation ==<br />
<br />
Parsers are typically evaluated with the LAS (Label Attachment Score) and UAS (Unlabeled Attachment Score) obtained by testing on a separate corpus. However, making an additional, separate corpus would be time consuming. To get around this, I divided the corpus into 10 roughly equal parts (10-11 sentences each). Then, I used each part as a test corpus for two parsers (with and without morphological information) which were trained on the remaining corpus. <br />
<br />
The partial corpora, conllu and udpipe files can all be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-eval bcl-eval directory] of the repository. The data from all the individual tests can be found in the [[#Appendix|Appendix]].<br />
<br />
To evaluate the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/dev parsers trained on the entire corpus], I took the average of all the UAS/LAS scores obtained on the 10 test sets, ± the standard deviation. The results are displayed below.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-withmorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|60.87 ± 5.67<br />
|50.00 ± 4.22<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-nomorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|50.51 ± 5.53<br />
|36.01 ± 6.59<br />
|}<br />
<br />
== Further Work ==<br />
<br />
Further work on this project would mainly involve increasing the size of the annotated corpus. My corpus of ~100 sentences and ~1200 tokens is very small in comparison to completed UD Treebanks, due to the time constraints of the project and the nature of the material I was working with (Wikipedia articles, no glosses). The most fully developed UD treebanks have 100-200K+ tokens annotated, which would yield much, much more accurate parsers.<br />
<br />
== Appendix ==<br />
<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation: Tests on 10 Subsets of Corpus<br />
|-<br />
! <br />
! UAS with morphology<br />
! LAS with morphology<br />
! UAS without morphology<br />
! LAS without morphology<br />
|-<br />
! Test 1<br />
| 55.37<br />
| 45.45<br />
| 48.76<br />
| 31.40<br />
|-<br />
! Test 2<br />
| 56.34<br />
| 46.48<br />
| 50.70<br />
| 38.03<br />
|-<br />
! Test 3<br />
| 64.29<br />
| 54.46<br />
| 57.14<br />
| 43.75<br />
|-<br />
! Test 4<br />
| 65.00<br />
| 49.17<br />
| 53.33<br />
| 36.67<br />
|-<br />
! Test 5<br />
| 56.44<br />
| 47.52<br />
| 40.59<br />
| 23.76<br />
|-<br />
! Test 6<br />
| 65.52<br />
| 52.41<br />
| 56.55<br />
| 38.62<br />
|-<br />
! Test 7<br />
| 52.89<br />
| 44.63<br />
| 46.28<br />
| 31.40<br />
|-<br />
! Test 8<br />
| 57.45<br />
| 48.94<br />
| 45.74<br />
| 34.04<br />
|-<br />
! Test 9<br />
| 67.65<br />
| 53.92<br />
| 49.02<br />
| 35.29<br />
|-<br />
! Test 10<br />
| 67.77<br />
| 57.02<br />
| 57.02<br />
| 47.11<br />
|}<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5290User:Eortici1/Final project2017-05-10T04:35:49Z<p>Eortici1: /* Introduction to UD */</p>
<hr />
<div>This page documents my final project, a larger [http://universaldependencies.org/ Universal Dependencies] corpus and annotation guidelines for Bikol.<br />
<br />
The <b><u>[https://github.swarthmore.edu/eortici1/UD_Bikol repository]</u></b> for this project is modeled closely after the repositories used by Universal Dependencies ([https://github.com/UniversalDependencies/UD_Greek example here]).<br />
<br />
== Introduction to UD ==<br />
<br />
The Universal Dependencies project was created to "provide a universal inventory of categories and guidelines to facilitate consistent annotation of similar constructions across languages". UD uses a set of typed dependency relations between words, where one word is the head of the sentence and all other words are dependent on another word in the sentence. These dependencies are generally universal and are meant to bring out cross-lingual typological parallels among the world's languages.<br />
<br />
For examples of the relations used in the Bikol treebank, see the [[#UD Annotation Guidelines|Annotation guidelines for Bikol]].<br />
<br />
<br />
For examples of annotated Bikol sentences, see [[#Corpus Information|the Bikol corpus]].<br />
<br />
== Corpus Information ==<br />
<br />
The annotated corpus can be found in [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-corpus.txt .txt] and .conllu ([https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-train.conllu with] and [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-train-nomorph.conllu without] morphological information) in the project repository.<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Tokens<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
<br />
The corpus contains extracts from the following sources:<br />
<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 08/05/2017<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
<br />
== Process ==<br />
<br />
In the following section, I address solutions to specific problems I encountered while annotating my corpus.<br />
<br />
=== Pluralization of nouns with 'mga' ===<br />
<br />
=== Linker in Modification Structures: 'na' ===<br />
<br />
<br />
* aux:amod, aux:nmod, aux:advmod, aux:nummod<br />
* check @dep and anything labeled !*<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
The final component of the project was to create annotation guidelines for the Bikol UD Treebank. This documentation would be useful to anyone who wished to contribute to the Treebank.<br />
<br />
Again, I used the same format that is used by the Universal Dependencies Project, using the [https://github.com/UniversalDependencies/docs/tree/pages-source/_sv-dep repository of Swedish guidelines] as a guide.<br />
<br />
The annotation guidelines can be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-dep bcl-dep directory] of the project's repository, and feature documentation on 25 of the most commonly used relations I used in my corpus, including the four language-specific relations I mentioned earlier in the [[#Process|Process section]].<br />
<br />
== Evaluation ==<br />
<br />
Parsers are typically evaluated with the LAS (Label Attachment Score) and UAS (Unlabeled Attachment Score) obtained by testing on a separate corpus. However, making an additional, separate corpus would be time consuming. To get around this, I divided the corpus into 10 roughly equal parts (10-11 sentences each). Then, I used each part as a test corpus for two parsers (with and without morphological information) which were trained on the remaining corpus. <br />
<br />
The partial corpora, conllu and udpipe files can all be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-eval bcl-eval directory] of the repository. The data from all the individual tests can be found in the [[#Appendix|Appendix]].<br />
<br />
To evaluate the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/dev parsers trained on the entire corpus], I took the average of all the UAS/LAS scores obtained on the 10 test sets, ± the standard deviation. The results are displayed below.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-withmorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|60.87 ± 5.67<br />
|50.00 ± 4.22<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-nomorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|50.51 ± 5.53<br />
|36.01 ± 6.59<br />
|}<br />
<br />
== Further Work ==<br />
<br />
Further work on this project would mainly involve increasing the size of the annotated corpus. My corpus of ~100 sentences and ~1200 tokens is very small in comparison to completed UD Treebanks, due to the time constraints of the project and the nature of the material I was working with (Wikipedia articles, no glosses). The most fully developed UD treebanks have 100-200K+ tokens annotated, which would yield much, much more accurate parsers.<br />
<br />
== Appendix ==<br />
<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation: Tests on 10 Subsets of Corpus<br />
|-<br />
! <br />
! UAS with morphology<br />
! LAS with morphology<br />
! UAS without morphology<br />
! LAS without morphology<br />
|-<br />
! Test 1<br />
| 55.37<br />
| 45.45<br />
| 48.76<br />
| 31.40<br />
|-<br />
! Test 2<br />
| 56.34<br />
| 46.48<br />
| 50.70<br />
| 38.03<br />
|-<br />
! Test 3<br />
| 64.29<br />
| 54.46<br />
| 57.14<br />
| 43.75<br />
|-<br />
! Test 4<br />
| 65.00<br />
| 49.17<br />
| 53.33<br />
| 36.67<br />
|-<br />
! Test 5<br />
| 56.44<br />
| 47.52<br />
| 40.59<br />
| 23.76<br />
|-<br />
! Test 6<br />
| 65.52<br />
| 52.41<br />
| 56.55<br />
| 38.62<br />
|-<br />
! Test 7<br />
| 52.89<br />
| 44.63<br />
| 46.28<br />
| 31.40<br />
|-<br />
! Test 8<br />
| 57.45<br />
| 48.94<br />
| 45.74<br />
| 34.04<br />
|-<br />
! Test 9<br />
| 67.65<br />
| 53.92<br />
| 49.02<br />
| 35.29<br />
|-<br />
! Test 10<br />
| 67.77<br />
| 57.02<br />
| 57.02<br />
| 47.11<br />
|}<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5289User:Eortici1/Final project2017-05-10T04:35:39Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger [http://universaldependencies.org/ Universal Dependencies] corpus and annotation guidelines for Bikol.<br />
<br />
The <b><u>[https://github.swarthmore.edu/eortici1/UD_Bikol repository]</u></b> for this project is modeled closely after the repositories used by Universal Dependencies ([https://github.com/UniversalDependencies/UD_Greek example here]).<br />
<br />
== Introduction to UD ==<br />
<br />
The Universal Dependencies project was created to "provide a universal inventory of categories and guidelines to facilitate consistent annotation of similar constructions across languages". UD uses a set of typed dependency relations between words, where one word is the head of the sentence and all other words are dependent on another word in the sentence. These dependencies are generally universal and are meant to bring out cross-lingual typological parallels among the world's languages.<br />
<br />
For examples of the relations used in the Bikol treebank, see the [[#UD Annotation Guidelines|Annotation guidelines for Bikol]].<br />
For examples of annotated Bikol sentences, see [[#Corpus Information|the Bikol corpus]].<br />
<br />
== Corpus Information ==<br />
<br />
The annotated corpus can be found in [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-corpus.txt .txt] and .conllu ([https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-train.conllu with] and [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-train-nomorph.conllu without] morphological information) in the project repository.<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Tokens<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
<br />
The corpus contains extracts from the following sources:<br />
<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 08/05/2017<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
<br />
== Process ==<br />
<br />
In the following section, I address solutions to specific problems I encountered while annotating my corpus.<br />
<br />
=== Pluralization of nouns with 'mga' ===<br />
<br />
=== Linker in Modification Structures: 'na' ===<br />
<br />
<br />
* aux:amod, aux:nmod, aux:advmod, aux:nummod<br />
* check @dep and anything labeled !*<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
The final component of the project was to create annotation guidelines for the Bikol UD Treebank. This documentation would be useful to anyone who wished to contribute to the Treebank.<br />
<br />
Again, I used the same format that is used by the Universal Dependencies Project, using the [https://github.com/UniversalDependencies/docs/tree/pages-source/_sv-dep repository of Swedish guidelines] as a guide.<br />
<br />
The annotation guidelines can be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-dep bcl-dep directory] of the project's repository, and feature documentation on 25 of the most commonly used relations I used in my corpus, including the four language-specific relations I mentioned earlier in the [[#Process|Process section]].<br />
<br />
== Evaluation ==<br />
<br />
Parsers are typically evaluated with the LAS (Label Attachment Score) and UAS (Unlabeled Attachment Score) obtained by testing on a separate corpus. However, making an additional, separate corpus would be time consuming. To get around this, I divided the corpus into 10 roughly equal parts (10-11 sentences each). Then, I used each part as a test corpus for two parsers (with and without morphological information) which were trained on the remaining corpus. <br />
<br />
The partial corpora, conllu and udpipe files can all be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-eval bcl-eval directory] of the repository. The data from all the individual tests can be found in the [[#Appendix|Appendix]].<br />
<br />
To evaluate the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/dev parsers trained on the entire corpus], I took the average of all the UAS/LAS scores obtained on the 10 test sets, ± the standard deviation. The results are displayed below.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-withmorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|60.87 ± 5.67<br />
|50.00 ± 4.22<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-nomorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|50.51 ± 5.53<br />
|36.01 ± 6.59<br />
|}<br />
<br />
== Further Work ==<br />
<br />
Further work on this project would mainly involve increasing the size of the annotated corpus. My corpus of ~100 sentences and ~1200 tokens is very small in comparison to completed UD Treebanks, due to the time constraints of the project and the nature of the material I was working with (Wikipedia articles, no glosses). The most fully developed UD treebanks have 100-200K+ tokens annotated, which would yield much, much more accurate parsers.<br />
<br />
== Appendix ==<br />
<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation: Tests on 10 Subsets of Corpus<br />
|-<br />
! <br />
! UAS with morphology<br />
! LAS with morphology<br />
! UAS without morphology<br />
! LAS without morphology<br />
|-<br />
! Test 1<br />
| 55.37<br />
| 45.45<br />
| 48.76<br />
| 31.40<br />
|-<br />
! Test 2<br />
| 56.34<br />
| 46.48<br />
| 50.70<br />
| 38.03<br />
|-<br />
! Test 3<br />
| 64.29<br />
| 54.46<br />
| 57.14<br />
| 43.75<br />
|-<br />
! Test 4<br />
| 65.00<br />
| 49.17<br />
| 53.33<br />
| 36.67<br />
|-<br />
! Test 5<br />
| 56.44<br />
| 47.52<br />
| 40.59<br />
| 23.76<br />
|-<br />
! Test 6<br />
| 65.52<br />
| 52.41<br />
| 56.55<br />
| 38.62<br />
|-<br />
! Test 7<br />
| 52.89<br />
| 44.63<br />
| 46.28<br />
| 31.40<br />
|-<br />
! Test 8<br />
| 57.45<br />
| 48.94<br />
| 45.74<br />
| 34.04<br />
|-<br />
! Test 9<br />
| 67.65<br />
| 53.92<br />
| 49.02<br />
| 35.29<br />
|-<br />
! Test 10<br />
| 67.77<br />
| 57.02<br />
| 57.02<br />
| 47.11<br />
|}<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5288User:Eortici1/Final project2017-05-10T04:28:31Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger [http://universaldependencies.org/ Universal Dependencies] corpus and annotation guidelines for Bikol.<br />
<br />
The <b><u>[https://github.swarthmore.edu/eortici1/UD_Bikol repository]</u></b> for this project is modeled closely after the repositories used by Universal Dependencies ([https://github.com/UniversalDependencies/UD_Greek example here]).<br />
<br />
== Corpus Information ==<br />
<br />
The annotated corpus can be found in [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-corpus.txt .txt] and .conllu ([https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-train.conllu with] and [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-train-nomorph.conllu without] morphological information) in the project repository.<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Tokens<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
<br />
The corpus contains extracts from the following sources:<br />
<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 08/05/2017<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
<br />
== Process ==<br />
<br />
In the following section, I address solutions to specific problems I encountered while annotating my corpus.<br />
<br />
=== Pluralization of nouns with 'mga' ===<br />
<br />
=== Linker in Modification Structures: 'na' ===<br />
<br />
<br />
* aux:amod, aux:nmod, aux:advmod, aux:nummod<br />
* check @dep and anything labeled !*<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
The final component of the project was to create annotation guidelines for the Bikol UD Treebank. This documentation would be useful to anyone who wished to contribute to the Treebank.<br />
<br />
Again, I used the same format that is used by the Universal Dependencies Project, using the [https://github.com/UniversalDependencies/docs/tree/pages-source/_sv-dep repository of Swedish guidelines] as a guide.<br />
<br />
The annotation guidelines can be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-dep bcl-dep directory] of the project's repository, and feature documentation on 25 of the most commonly used relations I used in my corpus, including the four language-specific relations I mentioned earlier in the [[#Process|Process section]].<br />
<br />
== Evaluation ==<br />
<br />
Parsers are typically evaluated with the LAS (Label Attachment Score) and UAS (Unlabeled Attachment Score) obtained by testing on a separate corpus. However, making an additional, separate corpus would be time consuming. To get around this, I divided the corpus into 10 roughly equal parts (10-11 sentences each). Then, I used each part as a test corpus for two parsers (with and without morphological information) which were trained on the remaining corpus. <br />
<br />
The partial corpora, conllu and udpipe files can all be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-eval bcl-eval directory] of the repository. The data from all the individual tests can be found in the [[#Appendix|Appendix]].<br />
<br />
To evaluate the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/dev parsers trained on the entire corpus], I took the average of all the UAS/LAS scores obtained on the 10 test sets, ± the standard deviation. The results are displayed below.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-withmorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|60.87 ± 5.67<br />
|50.00 ± 4.22<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-nomorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|50.51 ± 5.53<br />
|36.01 ± 6.59<br />
|}<br />
<br />
== Further Work ==<br />
<br />
Further work on this project would mainly involve increasing the size of the annotated corpus. My corpus of ~100 sentences and ~1200 tokens is very small in comparison to completed UD Treebanks, due to the time constraints of the project and the nature of the material I was working with (Wikipedia articles, no glosses). The most fully developed UD treebanks have 100-200K+ tokens annotated, which would yield much, much more accurate parsers.<br />
<br />
== Appendix ==<br />
<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation: Tests on 10 Subsets of Corpus<br />
|-<br />
! <br />
! UAS with morphology<br />
! LAS with morphology<br />
! UAS without morphology<br />
! LAS without morphology<br />
|-<br />
! Test 1<br />
| 55.37<br />
| 45.45<br />
| 48.76<br />
| 31.40<br />
|-<br />
! Test 2<br />
| 56.34<br />
| 46.48<br />
| 50.70<br />
| 38.03<br />
|-<br />
! Test 3<br />
| 64.29<br />
| 54.46<br />
| 57.14<br />
| 43.75<br />
|-<br />
! Test 4<br />
| 65.00<br />
| 49.17<br />
| 53.33<br />
| 36.67<br />
|-<br />
! Test 5<br />
| 56.44<br />
| 47.52<br />
| 40.59<br />
| 23.76<br />
|-<br />
! Test 6<br />
| 65.52<br />
| 52.41<br />
| 56.55<br />
| 38.62<br />
|-<br />
! Test 7<br />
| 52.89<br />
| 44.63<br />
| 46.28<br />
| 31.40<br />
|-<br />
! Test 8<br />
| 57.45<br />
| 48.94<br />
| 45.74<br />
| 34.04<br />
|-<br />
! Test 9<br />
| 67.65<br />
| 53.92<br />
| 49.02<br />
| 35.29<br />
|-<br />
! Test 10<br />
| 67.77<br />
| 57.02<br />
| 57.02<br />
| 47.11<br />
|}<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5287User:Eortici1/Final project2017-05-10T04:23:51Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
The <b><u>[https://github.swarthmore.edu/eortici1/UD_Bikol repository]</u></b> for this project is modeled closely after the repositories used by Universal Dependencies.<br />
<br />
Issues<br />
* links<br />
* mga<br />
<br />
Re-evaluate NA !!!<br />
* aux:amod, aux:nmod, aux:advmod, aux:nummod<br />
* check @dep and anything labeled !* <br />
<br />
<br />
== Corpus Information ==<br />
<br />
The annotated corpus can be found in [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-corpus.txt .txt] and .conllu ([https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-train.conllu with] and [https://github.swarthmore.edu/eortici1/UD_Bikol/blob/master/bcl-ud-train-nomorph.conllu without] morphological information) in the project repository.<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Tokens<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
<br />
The corpus contains extracts from the following sources:<br />
<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 08/05/2017<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
<br />
== Process ==<br />
<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
The final component of the project was to create annotation guidelines for the Bikol UD Treebank. This documentation would be useful to anyone who wished to contribute to the Treebank.<br />
<br />
Again, I used the same format that is used by the Universal Dependencies Project, using the [https://github.com/UniversalDependencies/docs/tree/pages-source/_sv-dep repository of Swedish guidelines] as a guide.<br />
<br />
The annotation guidelines can be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-dep bcl-dep directory] of the project's repository, and feature documentation on 25 of the most commonly used relations I used in my corpus, including the four language-specific relations I mentioned earlier in the [[#Process|Process section]].<br />
<br />
== Evaluation ==<br />
<br />
Parsers are typically evaluated with the LAS (Label Attachment Score) and UAS (Unlabeled Attachment Score) obtained by testing on a separate corpus. However, making an additional, separate corpus would be time consuming. To get around this, I divided the corpus into 10 roughly equal parts (10-11 sentences each). Then, I used each part as a test corpus for two parsers (with and without morphological information) which were trained on the remaining corpus. <br />
<br />
The partial corpora, conllu and udpipe files can all be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-eval bcl-eval directory] of the repository. The data from all the individual tests can be found in the [[#Appendix|Appendix]].<br />
<br />
To evaluate the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/dev parsers trained on the entire corpus], I took the average of all the UAS/LAS scores obtained on the 10 test sets, ± the standard deviation. The results are displayed below.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-withmorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|60.87 ± 5.67<br />
|50.00 ± 4.22<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-nomorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|50.51 ± 5.53<br />
|36.01 ± 6.59<br />
|}<br />
<br />
== Further Work ==<br />
<br />
Further work on this project would mainly involve increasing the size of the annotated corpus. My corpus of ~100 sentences and ~1200 tokens is very small in comparison to completed UD Treebanks, due to the time constraints of the project and the nature of the material I was working with (Wikipedia articles, no glosses). The most fully developed UD treebanks have 100-200K+ tokens annotated, which would yield much, much more accurate parsers.<br />
<br />
== Appendix ==<br />
<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation: Tests on 10 Subsets of Corpus<br />
|-<br />
! <br />
! UAS with morphology<br />
! LAS with morphology<br />
! UAS without morphology<br />
! LAS without morphology<br />
|-<br />
! Test 1<br />
| 55.37<br />
| 45.45<br />
| 48.76<br />
| 31.40<br />
|-<br />
! Test 2<br />
| 56.34<br />
| 46.48<br />
| 50.70<br />
| 38.03<br />
|-<br />
! Test 3<br />
| 64.29<br />
| 54.46<br />
| 57.14<br />
| 43.75<br />
|-<br />
! Test 4<br />
| 65.00<br />
| 49.17<br />
| 53.33<br />
| 36.67<br />
|-<br />
! Test 5<br />
| 56.44<br />
| 47.52<br />
| 40.59<br />
| 23.76<br />
|-<br />
! Test 6<br />
| 65.52<br />
| 52.41<br />
| 56.55<br />
| 38.62<br />
|-<br />
! Test 7<br />
| 52.89<br />
| 44.63<br />
| 46.28<br />
| 31.40<br />
|-<br />
! Test 8<br />
| 57.45<br />
| 48.94<br />
| 45.74<br />
| 34.04<br />
|-<br />
! Test 9<br />
| 67.65<br />
| 53.92<br />
| 49.02<br />
| 35.29<br />
|-<br />
! Test 10<br />
| 67.77<br />
| 57.02<br />
| 57.02<br />
| 47.11<br />
|}<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5286User:Eortici1/Final project2017-05-10T04:19:04Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
The <b><u>[https://github.swarthmore.edu/eortici1/UD_Bikol repository]</u></b> for this project is modeled closely after the repositories used by Universal Dependencies.<br />
<br />
Issues<br />
* links<br />
* mga<br />
<br />
Re-evaluate NA !!!<br />
* aux:amod, aux:nmod, aux:advmod, aux:nummod<br />
* check @dep and anything labeled !* <br />
<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Tokens<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 08/05/2017<br />
<br />
== Process ==<br />
<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
The final component of the project was to create annotation guidelines for the Bikol UD Treebank. This documentation would be useful to anyone who wished to contribute to the Treebank.<br />
<br />
Again, I used the same format that is used by the Universal Dependencies Project, using the [https://github.com/UniversalDependencies/docs/tree/pages-source/_sv-dep repository of Swedish guidelines] as a guide.<br />
<br />
The annotation guidelines can be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-dep bcl-dep directory] of the project's repository, and feature documentation on 25 of the most commonly used relations I used in my corpus, including the four language-specific relations I mentioned earlier in the [[#Process|Process section]].<br />
<br />
== Evaluation ==<br />
<br />
Parsers are typically evaluated with the LAS (Label Attachment Score) and UAS (Unlabeled Attachment Score) obtained by testing on a separate corpus. However, making an additional, separate corpus would be time consuming. To get around this, I divided the corpus into 10 roughly equal parts (10-11 sentences each). Then, I used each part as a test corpus for two parsers (with and without morphological information) which were trained on the remaining corpus. <br />
<br />
The partial corpora, conllu and udpipe files can all be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-eval bcl-eval directory] of the repository. The data from all the individual tests can be found in the [[#Appendix|Appendix]].<br />
<br />
To evaluate the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/dev parsers trained on the entire corpus], I took the average of all the UAS/LAS scores obtained on the 10 test sets, ± the standard deviation. The results are displayed below.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-withmorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|60.87 ± 5.67<br />
|50.00 ± 4.22<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-nomorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|50.51 ± 5.53<br />
|36.01 ± 6.59<br />
|}<br />
<br />
== Further Work ==<br />
<br />
Further work on this project would mainly involve increasing the size of the annotated corpus. My corpus of ~100 sentences and ~1200 tokens is very small in comparison to completed UD Treebanks, due to the time constraints of the project and the nature of the material I was working with (Wikipedia articles, no glosses). The most fully developed UD treebanks have 100-200K+ tokens annotated, which would yield much, much more accurate parsers.<br />
<br />
== Appendix ==<br />
<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation: Tests on 10 Subsets of Corpus<br />
|-<br />
! <br />
! UAS with morphology<br />
! LAS with morphology<br />
! UAS without morphology<br />
! LAS without morphology<br />
|-<br />
! Test 1<br />
| 55.37<br />
| 45.45<br />
| 48.76<br />
| 31.40<br />
|-<br />
! Test 2<br />
| 56.34<br />
| 46.48<br />
| 50.70<br />
| 38.03<br />
|-<br />
! Test 3<br />
| 64.29<br />
| 54.46<br />
| 57.14<br />
| 43.75<br />
|-<br />
! Test 4<br />
| 65.00<br />
| 49.17<br />
| 53.33<br />
| 36.67<br />
|-<br />
! Test 5<br />
| 56.44<br />
| 47.52<br />
| 40.59<br />
| 23.76<br />
|-<br />
! Test 6<br />
| 65.52<br />
| 52.41<br />
| 56.55<br />
| 38.62<br />
|-<br />
! Test 7<br />
| 52.89<br />
| 44.63<br />
| 46.28<br />
| 31.40<br />
|-<br />
! Test 8<br />
| 57.45<br />
| 48.94<br />
| 45.74<br />
| 34.04<br />
|-<br />
! Test 9<br />
| 67.65<br />
| 53.92<br />
| 49.02<br />
| 35.29<br />
|-<br />
! Test 10<br />
| 67.77<br />
| 57.02<br />
| 57.02<br />
| 47.11<br />
|}<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5285User:Eortici1/Final project2017-05-10T04:18:28Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
The <b><u>[https://github.swarthmore.edu/eortici1/UD_Bikol repository]</u></b> for this project is modeled closely after the repositories used by Universal Dependencies.<br />
<br />
Issues<br />
* links<br />
* mga<br />
<br />
Re-evaluate NA !!!<br />
* aux:amod, aux:nmod, aux:advmod, aux:nummod<br />
* check @dep and anything labeled !* <br />
<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Tokens<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 08/05/2017<br />
<br />
== Process ==<br />
<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
The final component of the project was to create annotation guidelines for the Bikol UD Treebank. This documentation would be useful to anyone who wished to contribute to the Treebank.<br />
<br />
Again, I used the same format that is used by the Universal Dependencies Project, using the [https://github.com/UniversalDependencies/docs/tree/pages-source/_sv-dep repository of Swedish guidelines] as a guide.<br />
<br />
The annotation guidelines can be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-dep bcl-dep directory] of the project's repository, and feature documentation on 25 of the most commonly used relations I used in my corpus, including the four language-specific relations I mentioned earlier in the [[#Process|Process section]].<br />
<br />
== Evaluation ==<br />
<br />
Parsers are typically evaluated with the LAS (Label Attachment Score) and UAS (Unlabeled Attachment Score) obtained by testing on a separate corpus. However, making an additional, separate corpus would be time consuming. To get around this, I divided the corpus into 10 roughly equal parts (10-11 sentences each). Then, I used each part as a test corpus for two parsers (with and without morphological information) which were trained on the remaining corpus. <br />
<br />
The partial corpora, conllu and udpipe files can all be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-eval bcl-eval directory] of the repository. The data from all the individual tests can be found in the Appendix of this Wiki page.<br />
<br />
To evaluate the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/dev parsers trained on the entire corpus], I took the average of all the UAS/LAS scores obtained on the 10 test sets, ± the standard deviation. The results are displayed below.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-withmorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|60.87 ± 5.67<br />
|50.00 ± 4.22<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-nomorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|50.51 ± 5.53<br />
|36.01 ± 6.59<br />
|}<br />
<br />
== Further Work ==<br />
<br />
Further work on this project would mainly involve increasing the size of the annotated corpus. My corpus of ~100 sentences and ~1200 tokens is very small in comparison to completed UD Treebanks, due to the time constraints of the project and the nature of the material I was working with (Wikipedia articles, no glosses). The most fully developed UD treebanks have 100-200K+ tokens annotated, which would yield much, much more accurate parsers.<br />
<br />
== Appendix ==<br />
<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation: Tests on 10 Subsets of Corpus<br />
|-<br />
! <br />
! UAS with morphology<br />
! LAS with morphology<br />
! UAS without morphology<br />
! LAS without morphology<br />
|-<br />
! Test 1<br />
| 55.37<br />
| 45.45<br />
| 48.76<br />
| 31.40<br />
|-<br />
! Test 2<br />
| 56.34<br />
| 46.48<br />
| 50.70<br />
| 38.03<br />
|-<br />
! Test 3<br />
| 64.29<br />
| 54.46<br />
| 57.14<br />
| 43.75<br />
|-<br />
! Test 4<br />
| 65.00<br />
| 49.17<br />
| 53.33<br />
| 36.67<br />
|-<br />
! Test 5<br />
| 56.44<br />
| 47.52<br />
| 40.59<br />
| 23.76<br />
|-<br />
! Test 6<br />
| 65.52<br />
| 52.41<br />
| 56.55<br />
| 38.62<br />
|-<br />
! Test 7<br />
| 52.89<br />
| 44.63<br />
| 46.28<br />
| 31.40<br />
|-<br />
! Test 8<br />
| 57.45<br />
| 48.94<br />
| 45.74<br />
| 34.04<br />
|-<br />
! Test 9<br />
| 67.65<br />
| 53.92<br />
| 49.02<br />
| 35.29<br />
|-<br />
! Test 10<br />
| 67.77<br />
| 57.02<br />
| 57.02<br />
| 47.11<br />
|}<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5284User:Eortici1/Final project2017-05-10T04:16:03Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
The <b><u>[https://github.swarthmore.edu/eortici1/UD_Bikol repository]</u></b> for this project is modeled closely after the repositories used by Universal Dependencies.<br />
<br />
Issues<br />
* links<br />
* mga<br />
<br />
Re-evaluate NA !!!<br />
* aux:amod, aux:nmod, aux:advmod, aux:nummod<br />
* check @dep and anything labeled !* <br />
<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Tokens<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 08/05/2017<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
The final component of the project was to create annotation guidelines for the Bikol UD Treebank. This documentation would be useful to anyone who wished to contribute to the Treebank.<br />
<br />
Again, I used the same format that is used by the Universal Dependencies Project, using the [https://github.com/UniversalDependencies/docs/tree/pages-source/_sv-dep repository of Swedish guidelines] as a guide.<br />
<br />
The annotation guidelines can be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-dep bcl-dep directory] of the project's repository, and feature documentation on 25 of the most commonly used relations I used in my corpus, including the four language-specific relations I mentioned earlier in the Development section.<br />
<br />
== Evaluation ==<br />
<br />
Parsers are typically evaluated with the LAS (Label Attachment Score) and UAS (Unlabeled Attachment Score) obtained by testing on a separate corpus. However, making an additional, separate corpus would be time consuming. To get around this, I divided the corpus into 10 roughly equal parts (10-11 sentences each). Then, I used each part as a test corpus for two parsers (with and without morphological information) which were trained on the remaining corpus. <br />
<br />
The partial corpora, conllu and udpipe files can all be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-eval bcl-eval directory] of the repository. The data from all the individual tests can be found in the Appendix of this Wiki page.<br />
<br />
To evaluate the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/dev parsers trained on the entire corpus], I took the average of all the UAS/LAS scores obtained on the 10 test sets, ± the standard deviation. The results are displayed below.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-withmorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|60.87 ± 5.67<br />
|50.00 ± 4.22<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-nomorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|50.51 ± 5.53<br />
|36.01 ± 6.59<br />
|}<br />
<br />
== Further Work ==<br />
<br />
Further work on this project would mainly involve increasing the size of the annotated corpus. My corpus of ~100 sentences and ~1200 tokens is very small in comparison to completed UD Treebanks, due to the time constraints of the project and the nature of the material I was working with (Wikipedia articles, no glosses). The most fully developed UD treebanks have 100-200K+ tokens annotated, which would yield much, much more accurate parsers.<br />
<br />
== Appendix ==<br />
<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation: Tests on 10 Subsets of Corpus<br />
|-<br />
! <br />
! UAS with morphology<br />
! LAS with morphology<br />
! UAS without morphology<br />
! LAS without morphology<br />
|-<br />
! Test 1<br />
| 55.37<br />
| 45.45<br />
| 48.76<br />
| 31.40<br />
|-<br />
! Test 2<br />
| 56.34<br />
| 46.48<br />
| 50.70<br />
| 38.03<br />
|-<br />
! Test 3<br />
| 64.29<br />
| 54.46<br />
| 57.14<br />
| 43.75<br />
|-<br />
! Test 4<br />
| 65.00<br />
| 49.17<br />
| 53.33<br />
| 36.67<br />
|-<br />
! Test 5<br />
| 56.44<br />
| 47.52<br />
| 40.59<br />
| 23.76<br />
|-<br />
! Test 6<br />
| 65.52<br />
| 52.41<br />
| 56.55<br />
| 38.62<br />
|-<br />
! Test 7<br />
| 52.89<br />
| 44.63<br />
| 46.28<br />
| 31.40<br />
|-<br />
! Test 8<br />
| 57.45<br />
| 48.94<br />
| 45.74<br />
| 34.04<br />
|-<br />
! Test 9<br />
| 67.65<br />
| 53.92<br />
| 49.02<br />
| 35.29<br />
|-<br />
! Test 10<br />
| 67.77<br />
| 57.02<br />
| 57.02<br />
| 47.11<br />
|}<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5280User:Eortici1/Final project2017-05-10T04:00:56Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
Issues<br />
* links<br />
* mga<br />
<br />
Re-evaluate NA !!!<br />
* aux:amod, aux:nmod, aux:advmod, aux:nummod<br />
* check @dep and anything labeled !* <br />
<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 08/05/2017<br />
<br />
== Evaluation ==<br />
<br />
Parsers are typically evaluated with the LAS (Label Attachment Score) and UAS (Unlabeled Attachment Score) obtained by testing on a separate corpus. However, making an additional, separate corpus would be time consuming. To get around this, I divided the corpus into 10 roughly equal parts (10-11 sentences each). Then, I used each part as a test corpus for two parsers (with and without morphological information) which were trained on the remaining corpus. <br />
<br />
The partial corpora, conllu and udpipe files can all be found in the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/bcl-eval bcl-eval directory] of the repository. The data from all the individual tests can be found in the Appendix of this Wiki page.<br />
<br />
To evaluate the [https://github.swarthmore.edu/eortici1/UD_Bikol/tree/master/dev parsers trained on the entire corpus], I took the average of all the UAS/LAS scores obtained on the 10 test sets, ± the standard deviation. The results are displayed below.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-withmorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|60.87 ± 5.67<br />
|50.00 ± 4.22<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-nomorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|50.51 ± 5.53<br />
|36.01 ± 6.59<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
== Appendix ==<br />
<br />
=== Evaluation ===<br />
<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation: Tests on 10 Subsets of Corpus<br />
|-<br />
! <br />
! UAS with morphology<br />
! LAS with morphology<br />
! UAS without morphology<br />
! LAS without morphology<br />
|-<br />
! Test 1<br />
| 55.37<br />
| 45.45<br />
| 48.76<br />
| 31.40<br />
|-<br />
! Test 2<br />
| 56.34<br />
| 46.48<br />
| 50.70<br />
| 38.03<br />
|-<br />
! Test 3<br />
| 64.29<br />
| 54.46<br />
| 57.14<br />
| 43.75<br />
|-<br />
! Test 4<br />
| 65.00<br />
| 49.17<br />
| 53.33<br />
| 36.67<br />
|-<br />
! Test 5<br />
| 56.44<br />
| 47.52<br />
| 40.59<br />
| 23.76<br />
|-<br />
! Test 6<br />
| 65.52<br />
| 52.41<br />
| 56.55<br />
| 38.62<br />
|-<br />
! Test 7<br />
| 52.89<br />
| 44.63<br />
| 46.28<br />
| 31.40<br />
|-<br />
! Test 8<br />
| 57.45<br />
| 48.94<br />
| 45.74<br />
| 34.04<br />
|-<br />
! Test 9<br />
| 67.65<br />
| 53.92<br />
| 49.02<br />
| 35.29<br />
|-<br />
! Test 10<br />
| 67.77<br />
| 57.02<br />
| 57.02<br />
| 47.11<br />
|}<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5279User:Eortici1/Final project2017-05-10T03:45:24Z<p>Eortici1: /* Evaluation */</p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
* [http://universaldependencies.org/new_language.html UD Procedures for new languages]<br />
<br />
Issues<br />
* links<br />
* mga<br />
<br />
Change sarong to a determiner instead of an nmod?<br />
<br />
Check notes on phone for some changes to be made.<br />
<br />
Re-evaluate NA !!!<br />
* aux:amod, aux:nmod, aux:advmod, aux:nummod<br />
* check @dep and anything labeled !* <br />
<br />
tr -cd @ < bcl.annotated.large.ud.txt | wc -c<br />
* counts number of @ in the file<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 10/05/2017<br />
<br />
== Evaluation ==<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-withmorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|60.87 ± 5.67<br />
|50.00 ± 4.22<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-nomorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|50.51 ± 5.53<br />
|36.01 ± 6.59<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
== Appendix ==<br />
<br />
=== Evaluation ===<br />
<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation: Tests on 10 Subsets of Corpus<br />
|-<br />
! <br />
! UAS with morphology<br />
! LAS with morphology<br />
! UAS without morphology<br />
! LAS without morphology<br />
|-<br />
! Test 1<br />
| 55.37<br />
| 45.45<br />
| 48.76<br />
| 31.40<br />
|-<br />
! Test 2<br />
| 56.34<br />
| 46.48<br />
| 50.70<br />
| 38.03<br />
|-<br />
! Test 3<br />
| 64.29<br />
| 54.46<br />
| 57.14<br />
| 43.75<br />
|-<br />
! Test 4<br />
| 65.00<br />
| 49.17<br />
| 53.33<br />
| 36.67<br />
|-<br />
! Test 5<br />
| 56.44<br />
| 47.52<br />
| 40.59<br />
| 23.76<br />
|-<br />
! Test 6<br />
| 65.52<br />
| 52.41<br />
| 56.55<br />
| 38.62<br />
|-<br />
! Test 7<br />
| 52.89<br />
| 44.63<br />
| 46.28<br />
| 31.40<br />
|-<br />
! Test 8<br />
| 57.45<br />
| 48.94<br />
| 45.74<br />
| 34.04<br />
|-<br />
! Test 9<br />
| 67.65<br />
| 53.92<br />
| 49.02<br />
| 35.29<br />
|-<br />
! Test 10<br />
| 67.77<br />
| 57.02<br />
| 57.02<br />
| 47.11<br />
|}<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5278User:Eortici1/Final project2017-05-10T03:44:30Z<p>Eortici1: /* Evaluation */</p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
* [http://universaldependencies.org/new_language.html UD Procedures for new languages]<br />
<br />
Issues<br />
* links<br />
* mga<br />
<br />
Change sarong to a determiner instead of an nmod?<br />
<br />
Check notes on phone for some changes to be made.<br />
<br />
Re-evaluate NA !!!<br />
* aux:amod, aux:nmod, aux:advmod, aux:nummod<br />
* check @dep and anything labeled !* <br />
<br />
tr -cd @ < bcl.annotated.large.ud.txt | wc -c<br />
* counts number of @ in the file<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 10/05/2017<br />
<br />
== Evaluation ==<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-withmorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|60.87 ± 5.67<br />
|50.00 ± 4.22<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl-nomorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|50.51 ± 5.53<br />
|36.01 ± 6.59<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
== Appendix ==<br />
<br />
=== Evaluation ===<br />
<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation<br />
|-<br />
! <br />
! UAS with morphology<br />
! LAS with morphology<br />
! UAS without morphology<br />
! LAS without morphology<br />
|-<br />
! Test 1<br />
| 55.37<br />
| 45.45<br />
| 48.76<br />
| 31.40<br />
|-<br />
! Test 2<br />
| 56.34<br />
| 46.48<br />
| 50.70<br />
| 38.03<br />
|-<br />
! Test 3<br />
| 64.29<br />
| 54.46<br />
| 57.14<br />
| 43.75<br />
|-<br />
! Test 4<br />
| 65.00<br />
| 49.17<br />
| 53.33<br />
| 36.67<br />
|-<br />
! Test 5<br />
| 56.44<br />
| 47.52<br />
| 40.59<br />
| 23.76<br />
|-<br />
! Test 6<br />
| 65.52<br />
| 52.41<br />
| 56.55<br />
| 38.62<br />
|-<br />
! Test 7<br />
| 52.89<br />
| 44.63<br />
| 46.28<br />
| 31.40<br />
|-<br />
! Test 8<br />
| 57.45<br />
| 48.94<br />
| 45.74<br />
| 34.04<br />
|-<br />
! Test 9<br />
| 67.65<br />
| 53.92<br />
| 49.02<br />
| 35.29<br />
|-<br />
! Test 10<br />
| 67.77<br />
| 57.02<br />
| 57.02<br />
| 47.11<br />
|}<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5277User:Eortici1/Final project2017-05-10T03:43:12Z<p>Eortici1: /* Evaluation */</p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
* [http://universaldependencies.org/new_language.html UD Procedures for new languages]<br />
<br />
Issues<br />
* links<br />
* mga<br />
<br />
Change sarong to a determiner instead of an nmod?<br />
<br />
Check notes on phone for some changes to be made.<br />
<br />
Re-evaluate NA !!!<br />
* aux:amod, aux:nmod, aux:advmod, aux:nummod<br />
* check @dep and anything labeled !* <br />
<br />
tr -cd @ < bcl.annotated.large.ud.txt | wc -c<br />
* counts number of @ in the file<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 10/05/2017<br />
<br />
== Evaluation ==<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.withmorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|60.87 ± 5.67<br />
|50.00 ± 4.22<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.nomorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|50.51 ± 5.53<br />
| ±<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
== Appendix ==<br />
<br />
=== Evaluation ===<br />
<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation<br />
|-<br />
! <br />
! UAS with morphology<br />
! LAS with morphology<br />
! UAS without morphology<br />
! LAS without morphology<br />
|-<br />
! Test 1<br />
| 55.37<br />
| 45.45<br />
| 48.76<br />
| 31.40<br />
|-<br />
! Test 2<br />
| 56.34<br />
| 46.48<br />
| 50.70<br />
| 38.03<br />
|-<br />
! Test 3<br />
| 64.29<br />
| 54.46<br />
| 57.14<br />
| 43.75<br />
|-<br />
! Test 4<br />
| 65.00<br />
| 49.17<br />
| 53.33<br />
| 36.67<br />
|-<br />
! Test 5<br />
| 56.44<br />
| 47.52<br />
| 40.59<br />
| 23.76<br />
|-<br />
! Test 6<br />
| 65.52<br />
| 52.41<br />
| 56.55<br />
| 38.62<br />
|-<br />
! Test 7<br />
| 52.89<br />
| 44.63<br />
| 46.28<br />
| 31.40<br />
|-<br />
! Test 8<br />
| 57.45<br />
| 48.94<br />
| 45.74<br />
| 34.04<br />
|-<br />
! Test 9<br />
| 67.65<br />
| 53.92<br />
| 49.02<br />
| 35.29<br />
|-<br />
! Test 10<br />
| 67.77<br />
| 57.02<br />
| 57.02<br />
| 47.11<br />
|}<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5276User:Eortici1/Final project2017-05-10T03:41:37Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
* [http://universaldependencies.org/new_language.html UD Procedures for new languages]<br />
<br />
Issues<br />
* links<br />
* mga<br />
<br />
Change sarong to a determiner instead of an nmod?<br />
<br />
Check notes on phone for some changes to be made.<br />
<br />
Re-evaluate NA !!!<br />
* aux:amod, aux:nmod, aux:advmod, aux:nummod<br />
* check @dep and anything labeled !* <br />
<br />
tr -cd @ < bcl.annotated.large.ud.txt | wc -c<br />
* counts number of @ in the file<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 10/05/2017<br />
<br />
== Evaluation ==<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.withmorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|60.87 ± 5.67<br />
|50.00 ± 4.22<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
== Appendix ==<br />
<br />
=== Evaluation ===<br />
<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation<br />
|-<br />
! <br />
! UAS with morphology<br />
! LAS with morphology<br />
! UAS without morphology<br />
! LAS without morphology<br />
|-<br />
! Test 1<br />
| 55.37<br />
| 45.45<br />
| 48.76<br />
| 31.40<br />
|-<br />
! Test 2<br />
| 56.34<br />
| 46.48<br />
| 50.70<br />
| 38.03<br />
|-<br />
! Test 3<br />
| 64.29<br />
| 54.46<br />
| 57.14<br />
| 43.75<br />
|-<br />
! Test 4<br />
| 65.00<br />
| 49.17<br />
| 53.33<br />
| 36.67<br />
|-<br />
! Test 5<br />
| 56.44<br />
| 47.52<br />
| 40.59<br />
| 23.76<br />
|-<br />
! Test 6<br />
| 65.52<br />
| 52.41<br />
| 56.55<br />
| 38.62<br />
|-<br />
! Test 7<br />
| 52.89<br />
| 44.63<br />
| 46.28<br />
| 31.40<br />
|-<br />
! Test 8<br />
| 57.45<br />
| 48.94<br />
| 45.74<br />
| 34.04<br />
|-<br />
! Test 9<br />
| 67.65<br />
| 53.92<br />
| 49.02<br />
| 35.29<br />
|-<br />
! Test 10<br />
| 67.77<br />
| 57.02<br />
| 57.02<br />
| 47.11<br />
|}<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5275User:Eortici1/Final project2017-05-10T03:40:30Z<p>Eortici1: /* Evaluation */</p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
* [http://universaldependencies.org/new_language.html UD Procedures for new languages]<br />
<br />
Issues<br />
* links<br />
* mga<br />
<br />
Change sarong to a determiner instead of an nmod?<br />
<br />
Check notes on phone for some changes to be made.<br />
<br />
Re-evaluate NA !!!<br />
* aux:amod, aux:nmod, aux:advmod, aux:nummod<br />
* check @dep and anything labeled !* <br />
<br />
tr -cd @ < bcl.annotated.large.ud.txt | wc -c<br />
* counts number of @ in the file<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 10/05/2017<br />
<br />
== Evaluation ==<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.withmorph.udpipe<br />
|-<br />
|UAS<br />
|LAS<br />
|-<br />
|60.87 ± 5.67<br />
|C2 UAS<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
== Appendix ==<br />
<br />
=== Evaluation ===<br />
<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation<br />
|-<br />
! <br />
! UAS with morphology<br />
! LAS with morphology<br />
! UAS without morphology<br />
! LAS without morphology<br />
|-<br />
! Test 1<br />
| 55.37<br />
| 45.45<br />
| 48.76<br />
| 31.40<br />
|-<br />
! Test 2<br />
| 56.34<br />
| 46.48<br />
| 50.70<br />
| 38.03<br />
|-<br />
! Test 3<br />
| 64.29<br />
| 54.46<br />
| 57.14<br />
| 43.75<br />
|-<br />
! Test 4<br />
| 65.00<br />
| 49.17<br />
| 53.33<br />
| 36.67<br />
|-<br />
! Test 5<br />
| 56.44<br />
| 47.52<br />
| 40.59<br />
| 23.76<br />
|-<br />
! Test 6<br />
| 65.52<br />
| 52.41<br />
| 56.55<br />
| 38.62<br />
|-<br />
! Test 7<br />
| 52.89<br />
| 44.63<br />
| 46.28<br />
| 31.40<br />
|-<br />
! Test 8<br />
| 57.45<br />
| 48.94<br />
| 45.74<br />
| 34.04<br />
|-<br />
! Test 9<br />
| 67.65<br />
| 53.92<br />
| 49.02<br />
| 35.29<br />
|-<br />
! Test 10<br />
| 67.77<br />
| 57.02<br />
| 57.02<br />
| 47.11<br />
|}<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5274User:Eortici1/Final project2017-05-10T03:36:28Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
* [http://universaldependencies.org/new_language.html UD Procedures for new languages]<br />
<br />
Issues<br />
* links<br />
* mga<br />
<br />
Change sarong to a determiner instead of an nmod?<br />
<br />
Check notes on phone for some changes to be made.<br />
<br />
Re-evaluate NA !!!<br />
* aux:amod, aux:nmod, aux:advmod, aux:nummod<br />
* check @dep and anything labeled !* <br />
<br />
tr -cd @ < bcl.annotated.large.ud.txt | wc -c<br />
* counts number of @ in the file<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 10/05/2017<br />
<br />
== Evaluation ==<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
== Appendix ==<br />
<br />
=== Evaluation ===<br />
<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation<br />
|-<br />
! <br />
! UAS with morphology<br />
! LAS with morphology<br />
! UAS without morphology<br />
! LAS without morphology<br />
|-<br />
! Test 1<br />
| 55.37<br />
| 45.45<br />
| 48.76<br />
| 31.40<br />
|-<br />
! Test 2<br />
| 56.34<br />
| 46.48<br />
| 50.70<br />
| 38.03<br />
|-<br />
! Test 3<br />
| 64.29<br />
| 54.46<br />
| 57.14<br />
| 43.75<br />
|-<br />
! Test 4<br />
| 65.00<br />
| 49.17<br />
| 53.33<br />
| 36.67<br />
|-<br />
! Test 5<br />
| 56.44<br />
| 47.52<br />
| 40.59<br />
| 23.76<br />
|-<br />
! Test 6<br />
| 65.52<br />
| 52.41<br />
| 56.55<br />
| 38.62<br />
|-<br />
! Test 7<br />
| 52.89<br />
| 44.63<br />
| 46.28<br />
| 31.40<br />
|-<br />
! Test 8<br />
| 57.45<br />
| 48.94<br />
| 45.74<br />
| 34.04<br />
|-<br />
! Test 9<br />
| 67.65<br />
| 53.92<br />
| 49.02<br />
| 35.29<br />
|-<br />
! Test 10<br />
| 67.77<br />
| 57.02<br />
| 57.02<br />
| 47.11<br />
|}<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5259User:Eortici1/Final project2017-05-09T23:04:43Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
* [http://universaldependencies.org/new_language.html UD Procedures for new languages]<br />
<br />
Issues<br />
* links<br />
* mga<br />
<br />
Change sarong to a determiner instead of an nmod?<br />
<br />
Check notes on phone for some changes to be made.<br />
<br />
Re-evaluate NA !!!<br />
* aux:amod, aux:nmod, aux:advmod, aux:nummod<br />
* check @dep and anything labeled !* <br />
<br />
tr -cd @ < bcl.annotated.large.ud.txt | wc -c<br />
* counts number of @ in the file<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|103<br />
|1163<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 10/05/2017<br />
<br />
== Evaluation ==<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5251User:Eortici1/Final project2017-05-09T20:31:36Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
* [http://universaldependencies.org/new_language.html UD Procedures for new languages]<br />
<br />
Issues<br />
* links<br />
* mga<br />
<br />
Change sarong to a determiner instead of an nmod?<br />
<br />
Check notes on phone for some changes to be made.<br />
<br />
Re-evaluate NA !!!<br />
* aux:amod, aux:nmod, aux:advmod, aux:nummod<br />
* check @dep and anything labeled !* <br />
<br />
tr -cd @ < bcl.annotated.large.ud.txt | wc -c<br />
* counts number of @ in the file<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|#sent<br />
|#forms<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 10/05/2017<br />
<br />
== Evaluation ==<br />
<br />
* Do I need to evaluate it on another separate corpus?<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5172User:Eortici1/Final project2017-05-09T02:42:05Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
* [http://universaldependencies.org/new_language.html UD Procedures for new languages]<br />
<br />
Issues<br />
* links<br />
* mga<br />
<br />
Change sarong to a determiner instead of an nmod?<br />
<br />
Check notes on phone for some changes to be made.<br />
<br />
Re-evaluate NA !!!<br />
<br />
tr -cd @ < bcl.annotated.large.ud.txt | wc -c<br />
* counts number of @ in the file<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|#sent<br />
|#forms<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 10/05/2017<br />
<br />
== Evaluation ==<br />
<br />
* Do I need to evaluate it on another separate corpus?<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5133User:Eortici1/Final project2017-05-07T20:08:30Z<p>Eortici1: /* Corpus Information */</p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
* [http://universaldependencies.org/new_language.html UD Procedures for new languages]<br />
<br />
Issues<br />
* links<br />
* mga<br />
<br />
Change sarong to a determiner instead of an nmod?<br />
<br />
<br />
tr -cd @ < bcl.annotated.large.ud.txt | wc -c<br />
* counts number of @ in the file<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|#sent<br />
|#forms<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles], accessed from 01/05/2017 to 10/05/2017<br />
<br />
== Evaluation ==<br />
<br />
* Do I need to evaluate it on another separate corpus?<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5132User:Eortici1/Final project2017-05-07T20:08:13Z<p>Eortici1: /* Corpus Information */</p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
* [http://universaldependencies.org/new_language.html UD Procedures for new languages]<br />
<br />
Issues<br />
* links<br />
* mga<br />
<br />
Change sarong to a determiner instead of an nmod?<br />
<br />
<br />
tr -cd @ < bcl.annotated.large.ud.txt | wc -c<br />
* counts number of @ in the file<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|#sent<br />
|#forms<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* [http://bcl.wikipedia.org Various Bikol Wikipedia articles accessed from 01/05/2017 to 10/05/2017]<br />
<br />
== Evaluation ==<br />
<br />
* Do I need to evaluate it on another separate corpus?<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5131User:Eortici1/Final project2017-05-07T20:07:59Z<p>Eortici1: /* Corpus Information */</p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
* [http://universaldependencies.org/new_language.html UD Procedures for new languages]<br />
<br />
Issues<br />
* links<br />
* mga<br />
<br />
Change sarong to a determiner instead of an nmod?<br />
<br />
<br />
tr -cd @ < bcl.annotated.large.ud.txt | wc -c<br />
* counts number of @ in the file<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|#sent<br />
|#forms<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* [bcl.wikipedia.org Various Bikol Wikipedia articles accessed from 01/05/2017 to 10/05/2017]<br />
<br />
== Evaluation ==<br />
<br />
* Do I need to evaluate it on another separate corpus?<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5130User:Eortici1/Final project2017-05-07T20:06:15Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
* [http://universaldependencies.org/new_language.html UD Procedures for new languages]<br />
<br />
Issues<br />
* links<br />
* mga<br />
<br />
Change sarong to a determiner instead of an nmod?<br />
<br />
<br />
tr -cd @ < bcl.annotated.large.ud.txt | wc -c<br />
* counts number of @ in the file<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|#sent<br />
|#forms<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* Add more sources here<br />
<br />
== Evaluation ==<br />
<br />
* Do I need to evaluate it on another separate corpus?<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=Bikol_and_Chamorro&diff=5017Bikol and Chamorro2017-05-03T22:43:37Z<p>Eortici1: </p>
<hr />
<div>Resources for machine translation between Bikol and Chamorro<br />
* [https://github.swarthmore.edu/ltarlin1/ling073-bcl-cha Final Bikol-Chamorro translation repository]<br />
* [https://wikis.swarthmore.edu/ling073/Bikol Bikol]<br />
* [https://wikis.swarthmore.edu/ling073/Chamorro Chamorro]<br />
<br />
== Developed Resources ==<br />
* [https://wikis.swarthmore.edu/ling073/Bikol_and_Chamorro/Contrastive_Grammar Contrastive Grammar]<br />
* [https://wikis.swarthmore.edu/ling073/Bikol_and_Chamorro/Lexical_Selection Lexical Selection]<br />
<br />
== Translation Repositories ==<br />
* [https://github.swarthmore.edu/ltarlin1/ling073-bcl-cha Bikol to Chamorro translation]<br />
* [https://github.swarthmore.edu/eortici1/ling073-cha-bcl Chamorro to Bikol translation]<br />
<br />
== Initial Evaluation ==<br />
=== bcl → cha Evalutaion ===<br />
* WER = 94.74%<br />
* PER = 84.21%<br />
* Proportion of stems translated correctly = 100.00%<br />
<br />
=== cha → bcl Evaluation ===<br />
* WER = 90.91%<br />
* PER = 86.36%<br />
* Proportion of stems translated correctly = 100.00%<br />
<br />
=== Lexical Selection ===<br />
Find at least two cases in each direction of a one-to-many mapping.<br />
<br />
==== Bikol-Chamorro ====<br />
===== bóla =====<br />
In Bikol, <b>bóla</b> means ball or compliment. In Chamorro, ball = <i>bola</i> and compliment = <i>komplimenta</i>.<br />
<br />
Lexical selection between bola and komplimenta is guided by the following rules:<br />
* Bola (ball) is chosen when:<br />
** Preceded by a verb such as throw, toss, catch, play, buy, borrow, lend, bounce<br />
** Followed by an adjective such as round, colors<br />
<br />
Example sentences<br />
* Bola (ball) is chosen in the following example:<br />
** Sublion mo si bola ki Jose (<u>Borrow</u> the <i>ball</i> from Jacob)<br />
* Komplimenta (compliment) is chosen in the following example:<br />
** <example here><br />
<br />
===== glóbo =====<br />
In Bikol, <b>glóbo</b> means map or balloon. In Chamorro, map = <i>mapa</i> and balloon = <i>abubu</i>.<br />
<br />
For now, the default selection is set to translate glóbo to abubu, or balloon in English.<br />
<br />
==== Chamorro-Bikol ====<br />
* In Chamorro, <b>guakse</b> means to scrub (clothes) or to turn inside out. In Bikol, to scrub (clothes) is <i>kuró-kusó</i> and to turn inside out is <i>balík</i>.<br />
* <b>maggem</b> means cautious or damp. In Bikol, <i>máan</i> means cautious and <i>basá</i> means damp.<br />
<br />
<br />
=== Structural Transfer ===<br />
*[https://wikis.swarthmore.edu/ling073/Bikol_and_Chamorro/Structural_transfer Structural Transfer wiki]<br />
<br />
== Final Evaluation ==<br />
<br />
=== Polished RBMT System ===<br />
<br />
==== Bcl-Cha ====<br />
===== Additional Stems =====<br />
* 100 additional stems (verbs, adjectives, and nouns) were added to the bilingual dictionaries and both monolingual dictionaries.<br />
===== Lexical Selection Rules =====<br />
* In Bikol, the adjective <i>sadít</i> can either mean (1) small, tiny; or (2) low (as a soft voice).<br />
** <i>Despasio</i> (low, soft) is chosen if the adjective <i>sadít</i> follows the noun <i>bóses</i>, or voice.<br />
** <i>Balaskes</i> (small, tiny) is the default and is chosen in all other circumstances.<br />
<br />
* In Bikol,<br />
** <i>Pasífiku</i> (Pacific) is chosen if the adjective <i>pasípiko</i> is in the exact phrase <i>dágat na pasípiko</i>, or Pacific Ocean. <br />
** <i>Mahgong</i> (peaceful) is the default and is chosen in all other circumstances.<br />
<br />
==== Bcl Transducer ====<br />
===== Disambiguation (x2) =====<br />
====== sa ======<br />
The Bikol word "sa" can have either of the following two meanings, illustrated by the following example sentences.<br />
* <u>Determiner</u>: the<br />
** Nagdumán akó sa haróng na dakúla. <i>(I went to the big house.)</i><br />
** Correct analysis: <code>^Nagdumán/dumán<vblex><past>$ ^akó/akó<prn><p1><sg>$ ^sa<det><def>$ ^haróng/haróng<n>$ ^na/na<link>$ ^dakúla<adj><sg>$^./.<sent>$</code><br />
* <u>Preposition</u>: to<br />
** Nagdumán akó sa Píli. <i>(I went to Píli.)</i><br />
** Correct analysis: <code>^Nagdumán/dumán<vblex><past>$ ^akó/akó<prn><p1><sg>$ ^sa/sa<pr>$ ^Píli/Píli<cog>$^./.<sent>$</code><br />
<br />
====== dakúla ======<br />
The Bikol word "dakúla" can have either of the following two meanings, illustrated by the following example sentences.<br />
* <u>Adjective</u>: big<br />
** Nagdumán akó sa haróng na dakúla. <i>(I went to the big house.)</i><br />
** Correct analysis: <code>^Nagdumán/dumán<vblex><past>$ ^akó/akó<prn><p1><sg>$ ^sa/sa<det><def>$ ^haróng/haróng<n>$ ^na/na<link>$ ^dakúla<adj><sg>$^./.<sent>$</code><br />
* <u>Verb</u>: to grow bigger<br />
** Dakúla an búrak. <i>(A flower grows.)</i><br />
** Correct analysis: <code>^dakúla/dakúla<vblex>$ ^an/an<det><ind>$ ^búrak/búrak<n>$^./.<sent>$</code><br />
<br />
<br />
<!--<br />
== List of extra words ==<br />
Nouns:<br />
*basúra: garbage<br />
*garáhe: garage <-------didnʼt have<br />
*matemátika: mathematics<br />
*dungó: nose<br />
*pasénsia: patience<br />
*bolsá: pocket<br />
*hílo: poison<br />
*lamésa: table<br />
*pábo: turkey<br />
*dága: knife<br />
*púso: heart<br />
*lalawgón: face<br />
*kolór: color<br />
*kapé: coffee<br />
*báybáy: coast<br />
*bádo: clothes<br />
*inápay: bread<br />
<br />
Verbs:<br />
*luwás: to exit, to leave<br />
*apód: to call (someone)<br />
*sublí: to borrow<br />
*hatód: to bring, to deliver<br />
*será: to close<br />
*kagrít: to scream<br />
*bakál: to shop <-------didnʼt have<br />
*tarám: to talk<br />
*halát: to wait<br />
*matá: to wake up<br />
*búhay: to live<br />
*taó: to give<br />
*dekorár: to decorate<br />
*bítbít: to carry<br />
<br />
== Other possible lexical selection options ==<br />
* bulan: moon, month (110)<br />
* dali (ma-): quickly, easily (123)<br />
* dampog: foggy, moist (124)<br />
* delikado: delicate, dangerous (129)<br />
* duros (ma-): windy, breezy; boastful *** guaifon, <br />
* gabat: heavy, grievous (154)<br />
* gana: enthusiasm, appetite (156)<br />
* grado: grade (temp.) and grade (1st grade) (164)<br />
gugu (165)<br />
gurano (167)<br />
gusto (168)<br />
gutos (169)<br />
<br />
== Notes for Evan ==<br />
* typo in first line of Test 1 - Chamorro word should be "linulí"<br />
* is the accent supposed to be over the > sign? Is that just a weird rendering thing that doesn't actually matter?<br />
* Test 2 line 2 - the root is actually "gumaʼ", weʼll add the suffix and vowel change in a later step<br />
* Test 3 line 1 - accent looks like it's in the wrong place on maéstro<br />
* Test 3 line 1 - again, the root is just "maʼestro" and the prefix will be added later<br />
* Test 5 line 1 - root is "metgot"<br />
* I put all the above words (except "garage" and "to shop") into the bcl-cha dix file. Youʼll have to put them in the cha-bcl file (unless you did already) the same way but switched left and right. We also have to add them to our respective lexc files eventually as well<br />
<br />
== Notes for Lee ==<br />
* Test 1, iinulí should be ulí<br />
* Test 4, nagdumán should be dumán<br />
* Test 5, makusóg should be kusóg<br />
<br />
--><br />
<br />
[[Category:Sp17_TranslationPairs]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5015User:Eortici1/Final project2017-05-03T03:26:16Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
* [http://universaldependencies.org/new_language.html UD Procedures for new languages]<br />
<br />
Issues<br />
* links<br />
* mga<br />
Search through T0D0<br />
<br />
<br />
tr -cd @ < bcl.annotated.large.ud.txt | wc -c<br />
* counts number of @ in the file<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|#sent<br />
|#forms<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* Add more sources here<br />
<br />
== Evaluation ==<br />
<br />
* Do I need to evaluate it on another separate corpus?<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5014User:Eortici1/Final project2017-05-03T03:26:07Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
* [http://universaldependencies.org/new_language.html UD Procedures for new languages]<br />
<br />
Issues<br />
* links<br />
* mga<br />
Search through T0D0<br />
tr -cd @ < bcl.annotated.large.ud.txt | wc -c<br />
* counts number of @ in the file<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|#sent<br />
|#forms<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* Add more sources here<br />
<br />
== Evaluation ==<br />
<br />
* Do I need to evaluate it on another separate corpus?<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5013User:Eortici1/Final project2017-05-03T01:39:01Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
* [http://universaldependencies.org/new_language.html UD Procedures for new languages]<br />
<br />
Issues<br />
* links<br />
* mga<br />
Search through T0D0<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|#sent<br />
|#forms<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* Add more sources here<br />
<br />
== Evaluation ==<br />
<br />
* Do I need to evaluate it on another separate corpus?<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=5008User:Eortici1/Final project2017-05-03T00:25:42Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
* [http://universaldependencies.org/new_language.html UD Procedures for new languages]<br />
<br />
Issues<br />
* links<br />
* mga<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|#sent<br />
|#forms<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* Add more sources here<br />
<br />
== Evaluation ==<br />
<br />
* Do I need to evaluate it on another separate corpus?<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=4990User:Eortici1/Final project2017-05-01T23:28:25Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
* [http://universaldependencies.org/new_language.html UD Procedures for new languages]<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|#sent<br />
|#forms<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* Add more sources here<br />
<br />
== Evaluation ==<br />
<br />
* Do I need to evaluate it on another separate corpus?<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=4989User:Eortici1/Final project2017-05-01T23:21:27Z<p>Eortici1: /* Corpus Information */</p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|#sent<br />
|#forms<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* Add more sources here<br />
<br />
== Evaluation ==<br />
<br />
* Do I need to evaluate it on another separate corpus?<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=4988User:Eortici1/Final project2017-05-01T23:20:59Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|32<br />
|198<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* Add more sources here<br />
<br />
== Evaluation ==<br />
<br />
* Do I need to evaluate it on another separate corpus?<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.large.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|C1 LAS<br />
|C1 UAS<br />
|-<br />
|Corpus 2<br />
|C2 LAS<br />
|C2 UAS<br />
|}<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=4987User:Eortici1/Final project2017-05-01T23:19:10Z<p>Eortici1: /* Corpus Information */</p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
== Corpus Information ==<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|Sentences<br />
|Forms<br />
|-<br />
|32<br />
|198<br />
|}<br />
<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* Add more sources here<br />
<br />
== Evaluation ==<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=4986User:Eortici1/Final project2017-05-01T23:17:40Z<p>Eortici1: </p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
== Corpus Information ==<br />
This corpus contains extracts from the following sources:<br />
* [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=bkl Universal Declaration of Human Rights] (used [http://www.ohchr.org/EN/UDHR/Pages/Language.aspx?LangID=eng English translation] for glosses)<br />
* Add more sources here<br />
<br />
== Evaluation ==<br />
<br />
== UD Annotation Guidelines ==<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=User:Eortici1/Final_project&diff=4985User:Eortici1/Final project2017-05-01T23:02:54Z<p>Eortici1: Created page with "This page documents my final project, a larger UD corpus and annotation guidelines for Bikol. == Corpus Information == == Evaluation == == UD Annotation Guidelines == Cate..."</p>
<hr />
<div>This page documents my final project, a larger UD corpus and annotation guidelines for Bikol.<br />
<br />
== Corpus Information ==<br />
== Evaluation ==<br />
== UD Annotation Guidelines ==<br />
<br />
[[Category:sp17_FinalProjects]] [[Category:Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=Bikol/Universal_Dependencies&diff=4984Bikol/Universal Dependencies2017-05-01T22:55:34Z<p>Eortici1: </p>
<hr />
<div>This page documents Bikol Universal Dependencies. The corpora and parsers used for this assignment can be found in [https://github.swarthmore.edu/eortici1/ling073-bcl this GitHub repository].<br />
<br />
== Evaluation ==<br />
<br />
This assignment involved two Bikol corpora. The first corpus, and the larger of the two, was used to train two parsers ([https://github.swarthmore.edu/eortici1/ling073-bcl/blob/master/dev/bcl.withmorph.pipe withmorph] and [https://github.swarthmore.edu/eortici1/ling073-bcl/blob/master/dev/bcl.nomorph.pipe nomorph]). Then, each of the two parsers were evaluated using both corpora, [https://github.swarthmore.edu/eortici1/ling073-bcl/blob/master/bcl.annotated.ud.conllu Corpus 1] and a smaller [https://github.swarthmore.edu/eortici1/ling073-bcl/blob/master/bcl.annotated2.ud.conllu Corpus 2].<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|<br />
|Sentences<br />
|Forms<br />
|-<br />
|Corpus 1<br />
|32<br />
|198<br />
|-<br />
|Corpus 2<br />
|16<br />
|111<br />
|}<br />
<br />
The following are the LAS (Label Attachment Score) and UAS (Unlabeled Attachment Score) results when evaluating the parser trained with morphological information (bcl.withmorph.udpipe) using Corpus 1 and Corpus 2.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|71.21%<br />
|80.81%<br />
|-<br />
|Corpus 2<br />
|47.75%<br />
|67.57%<br />
|}<br />
<br />
The following are the LAS and UAS results when evaluating the parser trained with morphological information removed (bcl.nomorph.udpipe) using Corpus 1 and Corpus 2.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|56.06%<br />
|68.18%<br />
|-<br />
|Corpus 2<br />
|36.04%<br />
|54.05%<br />
|}<br />
<br />
== Dependency relations ==<br />
The following five sections illustrate examples of dependency relations used frequently in Bikol.<br />
<br />
=== advcl ===<br />
According to the Universal Dependencies documentation, an adverbial clause modifier is a clause which modifies a verb or other predicate as a modifier and not as a core complement. In my corpora, this relation came up in if statements.<br />
<br />
<pre><br />
"<Kon>"<br />
"kon" cnjsub @advmod #1->2<br />
"<makuá>"<br />
"kuá" vblex inf @advcl #2->5<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #3->2<br />
"<,>"<br />
"," cm @punct #4->5<br />
"<itaó>"<br />
"taó" vblex inf @root #5->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #6->5<br />
"<saná>"<br />
"saná" adv @advmod #7->5<br />
"<sakó>"<br />
"sakó" prn p1 sg @iobj #8->5<br />
"<.>"<br />
"." sent @punct #9->5<br />
</pre><br />
* This sentence translates it to "If you find it, just give it to me." The first clause, "If you find it", is an adverbial clause modifier which modifies "just give it to me". Thus, the root of the adverbial clause modifier, makuá, has an advcl dependency on the root of the independent clause that it modifies, itaó.<br />
<br />
<pre><br />
"<Kon>"<br />
"Kon" cnjsub @advmod #1->2<br />
"<maglúto>"<br />
"lúto" vblex inf @advcl #2->7<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #3->2<br />
"<nin>"<br />
"nin" link @clf #4->5<br />
"<jéllo>"<br />
"jéllo" n @obj #5->2<br />
"<,>"<br />
"," cm @punct #6->7<br />
"<magkuá>"<br />
"kuá" vblex inf @root #7->0<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #8->7<br />
"<nin>"<br />
"nin" link @clf #9->10<br />
"<pláto>"<br />
"pláto" n @obj #10->7<br />
"<sagkód>"<br />
"sagkód" adj sg @cc #11->12 ! and<br />
"<kutsára>"<br />
"kutsára" n @conj #12->10<br />
"<.>"<br />
"." sent @punct #13->2<br />
</pre><br />
* This sentence essentially has the same structure as the previous example. The root of the adverbial clause modifier, maglúto, has an advcl dependency on magkuá, the root of the independent clause.<br />
<br />
=== obl ===<br />
The obl (oblique) relation is used for a nominal that serves as a non-core argument or adjunct. In Bikol, this typically means that it functions as an adverbial attached to a predicate.<br />
<br />
<pre><br />
"<Hanápon>"<br />
"hánap" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<siyá>"<br />
"siyá" prn p3 sg @obj #3->1<br />
"<sa>"<br />
"sa" pr @case #4->5<br />
"<cafeteria>"<br />
"cafeteria" n @obl #5->1<br />
"<.>"<br />
"." sent @punct #6->1<br />
</pre><br />
* In this sentence, meaning "Look for him/her <i>in the cafeteria</i>", cafeteria is a nominal serving as a non-core argument and modifying the verb, hanápon (to look). Therefore cafeteria has an obl relation to hanápon.<br />
<br />
<pre><br />
"<Sublión>"<br />
"sublí" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<si>"<br />
"si" det def @det #3->4<br />
"<bóla>"<br />
"bóla" n @obj #4->1<br />
"<ki>"<br />
"ki" pr @case #5->6<br />
"<Jacób>"<br />
"Jacób" cog @obl #6->1<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
*This sentence behaves similarly as the previous example, except now the nominal in the oblique relation is a proper name (Jacób). It means "Borrow the ball from Jacob".<br />
<br />
=== amod ===<br />
The amod relation refers to adjectival modifier, which simply modifies the meaning of a noun.<br />
<br />
<pre><br />
"<Nagdumán>"<br />
"dumán" vblex past @root #1->0<br />
"<akó>"<br />
"akó" prn p1 sg @nsubj #2->1<br />
"<sa>"<br />
"sa" det def @det #3->4<br />
"<haróng>"<br />
"haróng" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<dakúla>"<br />
"dakúla" adj sg @amod #6->4<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
* Translating to "I left the big house", this sentence demonstrates the most common structure of adjectival modifiers in Bikol, where the noun is followed by a linking word and the adjective. Dakúla (big), has an amod relation to the noun haróng (house).<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Although syntactically this sentence is very similar to the previous example, with a noun followed by a linking word and an adjective, this demonstrates a different kind of adjectival modifier. The adjective "nakadukót" is derived from the verb stem "dukót", meaning to glue or attach. The "naka" prefix changes the word into an adjective, meaning attached or stuck on. The sentence translates to, "Remove the poster attached to the bridge."<br />
<br />
=== obj ===<br />
The object relation relates a verb and its object. Bikol is a VSO language, so the object of the verb is usually found after the subject, like in the following two examples.<br />
<br />
<pre><br />
"<Iinulí>"<br />
"ulí" vblex pres @root #1->0<br />
"<ko>"<br />
"ko" prn p1 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<libró>"<br />
"libró" n @obj #4->1<br />
"<.>"<br />
"." sent @punct #5->1<br />
</pre><br />
* In this sentence, which translates to "I return a book", libró is the object and depends on the verb, iinulí.<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Similarly, pósters is the object of this sentence and depends on the verb, ipaháli.<br />
<br />
=== advmod ===<br />
An adverbial modifier of a word is an adverb or adverbial phrase that modifies a predicate or modifier word.<br />
<br />
<pre><br />
"<Patosón>"<br />
"patós" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pansít>"<br />
"pansít" n @obj #4->1<br />
"<,>"<br />
"," cm @punct #5->6<br />
"<tápos>"<br />
"tápos" cnjadv @conj #6->1<br />
"<,>"<br />
"," cm @punct #7->8<br />
"<darahón>"<br />
"dará" vblex inf @conj #8->1<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #9->8<br />
"<digdí>"<br />
"digdí" adv @advmod #10->8<br />
"<.>"<br />
"." sent @punct #11->1<br />
</pre><br />
* In this sentence, digdí (or here) is the adverbial modifier that modifies the verb darahón (the command form of 'bring'). <br />
<br />
<pre><br />
"<Tápos>"<br />
"tápos" adj sg @root #1->0<br />
"<na>"<br />
"na" adv @advmod #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<prográma>"<br />
"prográma" n @nsubj #4->1<br />
"<sa>"<br />
"sa" pr @case #5->6<br />
"<gym>"<br />
"gym" n @nmod #6->4<br />
"<?>"<br />
"?" sent @punct #7->1<br />
</pre><br />
*This sentence translates to "Is the program in the gym already done?" The predicate is an implied "is", so the root the adjective, tápos (done). The adverbial modifier na (already) describes this adjective.<br />
<br />
<br />
[[Category: sp17_UD]]<br />
[[Category: Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=Bikol/Universal_Dependencies&diff=4983Bikol/Universal Dependencies2017-05-01T22:49:00Z<p>Eortici1: /* Evaluation */</p>
<hr />
<div>This page documents Bikol Universal Dependencies. <br />
<br />
== Evaluation ==<br />
<br />
This assignment involved two Bikol corpora. The first corpus, and the larger of the two, was used to train two parsers (withmorph and nomorph). Then, each of the two parsers were evaluated using both corpora, Corpus 1 and a smaller Corpus 2.<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|<br />
|Sentences<br />
|Forms<br />
|-<br />
|Corpus 1<br />
|32<br />
|198<br />
|-<br />
|Corpus 2<br />
|16<br />
|111<br />
|}<br />
<br />
The following are the LAS (Label Attachment Score) and UAS (Unlabeled Attachment Score) results when evaluating the parser trained with morphological information (bcl.withmorph.udpipe) using Corpus 1 and Corpus 2.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|71.21%<br />
|80.81%<br />
|-<br />
|Corpus 2<br />
|47.75%<br />
|67.57%<br />
|}<br />
<br />
The following are the LAS and UAS results when evaluating the parser trained with morphological information removed (bcl.nomorph.udpipe) using Corpus 1 and Corpus 2.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|56.06%<br />
|68.18%<br />
|-<br />
|Corpus 2<br />
|36.04%<br />
|54.05%<br />
|}<br />
<br />
== Dependency relations ==<br />
The following five sections illustrate examples of dependency relations used frequently in Bikol.<br />
<br />
=== advcl ===<br />
According to the Universal Dependencies documentation, an adverbial clause modifier is a clause which modifies a verb or other predicate as a modifier and not as a core complement. In my corpora, this relation came up in if statements.<br />
<br />
<pre><br />
"<Kon>"<br />
"kon" cnjsub @advmod #1->2<br />
"<makuá>"<br />
"kuá" vblex inf @advcl #2->5<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #3->2<br />
"<,>"<br />
"," cm @punct #4->5<br />
"<itaó>"<br />
"taó" vblex inf @root #5->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #6->5<br />
"<saná>"<br />
"saná" adv @advmod #7->5<br />
"<sakó>"<br />
"sakó" prn p1 sg @iobj #8->5<br />
"<.>"<br />
"." sent @punct #9->5<br />
</pre><br />
* This sentence translates it to "If you find it, just give it to me." The first clause, "If you find it", is an adverbial clause modifier which modifies "just give it to me". Thus, the root of the adverbial clause modifier, makuá, has an advcl dependency on the root of the independent clause that it modifies, itaó.<br />
<br />
<pre><br />
"<Kon>"<br />
"Kon" cnjsub @advmod #1->2<br />
"<maglúto>"<br />
"lúto" vblex inf @advcl #2->7<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #3->2<br />
"<nin>"<br />
"nin" link @clf #4->5<br />
"<jéllo>"<br />
"jéllo" n @obj #5->2<br />
"<,>"<br />
"," cm @punct #6->7<br />
"<magkuá>"<br />
"kuá" vblex inf @root #7->0<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #8->7<br />
"<nin>"<br />
"nin" link @clf #9->10<br />
"<pláto>"<br />
"pláto" n @obj #10->7<br />
"<sagkód>"<br />
"sagkód" adj sg @cc #11->12 ! and<br />
"<kutsára>"<br />
"kutsára" n @conj #12->10<br />
"<.>"<br />
"." sent @punct #13->2<br />
</pre><br />
* This sentence essentially has the same structure as the previous example. The root of the adverbial clause modifier, maglúto, has an advcl dependency on magkuá, the root of the independent clause.<br />
<br />
=== obl ===<br />
The obl (oblique) relation is used for a nominal that serves as a non-core argument or adjunct. In Bikol, this typically means that it functions as an adverbial attached to a predicate.<br />
<br />
<pre><br />
"<Hanápon>"<br />
"hánap" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<siyá>"<br />
"siyá" prn p3 sg @obj #3->1<br />
"<sa>"<br />
"sa" pr @case #4->5<br />
"<cafeteria>"<br />
"cafeteria" n @obl #5->1<br />
"<.>"<br />
"." sent @punct #6->1<br />
</pre><br />
* In this sentence, meaning "Look for him/her <i>in the cafeteria</i>", cafeteria is a nominal serving as a non-core argument and modifying the verb, hanápon (to look). Therefore cafeteria has an obl relation to hanápon.<br />
<br />
<pre><br />
"<Sublión>"<br />
"sublí" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<si>"<br />
"si" det def @det #3->4<br />
"<bóla>"<br />
"bóla" n @obj #4->1<br />
"<ki>"<br />
"ki" pr @case #5->6<br />
"<Jacób>"<br />
"Jacób" cog @obl #6->1<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
*This sentence behaves similarly as the previous example, except now the nominal in the oblique relation is a proper name (Jacób). It means "Borrow the ball from Jacob".<br />
<br />
=== amod ===<br />
The amod relation refers to adjectival modifier, which simply modifies the meaning of a noun.<br />
<br />
<pre><br />
"<Nagdumán>"<br />
"dumán" vblex past @root #1->0<br />
"<akó>"<br />
"akó" prn p1 sg @nsubj #2->1<br />
"<sa>"<br />
"sa" det def @det #3->4<br />
"<haróng>"<br />
"haróng" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<dakúla>"<br />
"dakúla" adj sg @amod #6->4<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
* Translating to "I left the big house", this sentence demonstrates the most common structure of adjectival modifiers in Bikol, where the noun is followed by a linking word and the adjective. Dakúla (big), has an amod relation to the noun haróng (house).<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Although syntactically this sentence is very similar to the previous example, with a noun followed by a linking word and an adjective, this demonstrates a different kind of adjectival modifier. The adjective "nakadukót" is derived from the verb stem "dukót", meaning to glue or attach. The "naka" prefix changes the word into an adjective, meaning attached or stuck on. The sentence translates to, "Remove the poster attached to the bridge."<br />
<br />
=== obj ===<br />
The object relation relates a verb and its object. Bikol is a VSO language, so the object of the verb is usually found after the subject, like in the following two examples.<br />
<br />
<pre><br />
"<Iinulí>"<br />
"ulí" vblex pres @root #1->0<br />
"<ko>"<br />
"ko" prn p1 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<libró>"<br />
"libró" n @obj #4->1<br />
"<.>"<br />
"." sent @punct #5->1<br />
</pre><br />
* In this sentence, which translates to "I return a book", libró is the object and depends on the verb, iinulí.<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Similarly, pósters is the object of this sentence and depends on the verb, ipaháli.<br />
<br />
=== advmod ===<br />
An adverbial modifier of a word is an adverb or adverbial phrase that modifies a predicate or modifier word.<br />
<br />
<pre><br />
"<Patosón>"<br />
"patós" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pansít>"<br />
"pansít" n @obj #4->1<br />
"<,>"<br />
"," cm @punct #5->6<br />
"<tápos>"<br />
"tápos" cnjadv @conj #6->1<br />
"<,>"<br />
"," cm @punct #7->8<br />
"<darahón>"<br />
"dará" vblex inf @conj #8->1<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #9->8<br />
"<digdí>"<br />
"digdí" adv @advmod #10->8<br />
"<.>"<br />
"." sent @punct #11->1<br />
</pre><br />
* In this sentence, digdí (or here) is the adverbial modifier that modifies the verb darahón (the command form of 'bring'). <br />
<br />
<pre><br />
"<Tápos>"<br />
"tápos" adj sg @root #1->0<br />
"<na>"<br />
"na" adv @advmod #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<prográma>"<br />
"prográma" n @nsubj #4->1<br />
"<sa>"<br />
"sa" pr @case #5->6<br />
"<gym>"<br />
"gym" n @nmod #6->4<br />
"<?>"<br />
"?" sent @punct #7->1<br />
</pre><br />
*This sentence translates to "Is the program in the gym already done?" The predicate is an implied "is", so the root the adjective, tápos (done). The adverbial modifier na (already) describes this adjective.<br />
<br />
<br />
[[Category: sp17_UD]]<br />
[[Category: Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=Bikol/Universal_Dependencies&diff=4982Bikol/Universal Dependencies2017-05-01T22:43:31Z<p>Eortici1: /* Evaluation */</p>
<hr />
<div>This page documents Bikol Universal Dependencies. <br />
<br />
== Evaluation ==<br />
<br />
This assignment involved two Bikol corpora. The first corpus, and the larger of the two, was used to train two parsers (withmorph and nomorph). Then, each of the two parsers were evaluated using both corpora, Corpus 1 and a smaller Corpus 2.<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|<br />
|Sentences<br />
|Forms<br />
|-<br />
|Corpus 1<br />
|32<br />
|198<br />
|-<br />
|Corpus 2<br />
|16<br />
|111<br />
|}<br />
<br />
The following are the LAS (Label Attachment Score) and UAS (Unlabeled Attachment Score) results when evaluating the parser trained with morphological information (bcl.withmorph.udpipe) using Corpus 1 and Corpus 2.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|71.21%<br />
|80.81%<br />
|-<br />
|Corpus 2<br />
|47.75%<br />
|67.57%<br />
|}<br />
<br />
The following are the LAS and UAS results when evaluating the parser trained with morphological information removed (bcl.nomorph.udpipe) using Corpus 1 and Corpus 2.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|[C1 LAS]<br />
|[C2 UAS]<br />
|-<br />
|Corpus 2<br />
|[C2 LAS]<br />
|[C2 UAS]<br />
|}<br />
<br />
== Dependency relations ==<br />
The following five sections illustrate examples of dependency relations used frequently in Bikol.<br />
<br />
=== advcl ===<br />
According to the Universal Dependencies documentation, an adverbial clause modifier is a clause which modifies a verb or other predicate as a modifier and not as a core complement. In my corpora, this relation came up in if statements.<br />
<br />
<pre><br />
"<Kon>"<br />
"kon" cnjsub @advmod #1->2<br />
"<makuá>"<br />
"kuá" vblex inf @advcl #2->5<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #3->2<br />
"<,>"<br />
"," cm @punct #4->5<br />
"<itaó>"<br />
"taó" vblex inf @root #5->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #6->5<br />
"<saná>"<br />
"saná" adv @advmod #7->5<br />
"<sakó>"<br />
"sakó" prn p1 sg @iobj #8->5<br />
"<.>"<br />
"." sent @punct #9->5<br />
</pre><br />
* This sentence translates it to "If you find it, just give it to me." The first clause, "If you find it", is an adverbial clause modifier which modifies "just give it to me". Thus, the root of the adverbial clause modifier, makuá, has an advcl dependency on the root of the independent clause that it modifies, itaó.<br />
<br />
<pre><br />
"<Kon>"<br />
"Kon" cnjsub @advmod #1->2<br />
"<maglúto>"<br />
"lúto" vblex inf @advcl #2->7<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #3->2<br />
"<nin>"<br />
"nin" link @clf #4->5<br />
"<jéllo>"<br />
"jéllo" n @obj #5->2<br />
"<,>"<br />
"," cm @punct #6->7<br />
"<magkuá>"<br />
"kuá" vblex inf @root #7->0<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #8->7<br />
"<nin>"<br />
"nin" link @clf #9->10<br />
"<pláto>"<br />
"pláto" n @obj #10->7<br />
"<sagkód>"<br />
"sagkód" adj sg @cc #11->12 ! and<br />
"<kutsára>"<br />
"kutsára" n @conj #12->10<br />
"<.>"<br />
"." sent @punct #13->2<br />
</pre><br />
* This sentence essentially has the same structure as the previous example. The root of the adverbial clause modifier, maglúto, has an advcl dependency on magkuá, the root of the independent clause.<br />
<br />
=== obl ===<br />
The obl (oblique) relation is used for a nominal that serves as a non-core argument or adjunct. In Bikol, this typically means that it functions as an adverbial attached to a predicate.<br />
<br />
<pre><br />
"<Hanápon>"<br />
"hánap" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<siyá>"<br />
"siyá" prn p3 sg @obj #3->1<br />
"<sa>"<br />
"sa" pr @case #4->5<br />
"<cafeteria>"<br />
"cafeteria" n @obl #5->1<br />
"<.>"<br />
"." sent @punct #6->1<br />
</pre><br />
* In this sentence, meaning "Look for him/her <i>in the cafeteria</i>", cafeteria is a nominal serving as a non-core argument and modifying the verb, hanápon (to look). Therefore cafeteria has an obl relation to hanápon.<br />
<br />
<pre><br />
"<Sublión>"<br />
"sublí" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<si>"<br />
"si" det def @det #3->4<br />
"<bóla>"<br />
"bóla" n @obj #4->1<br />
"<ki>"<br />
"ki" pr @case #5->6<br />
"<Jacób>"<br />
"Jacób" cog @obl #6->1<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
*This sentence behaves similarly as the previous example, except now the nominal in the oblique relation is a proper name (Jacób). It means "Borrow the ball from Jacob".<br />
<br />
=== amod ===<br />
The amod relation refers to adjectival modifier, which simply modifies the meaning of a noun.<br />
<br />
<pre><br />
"<Nagdumán>"<br />
"dumán" vblex past @root #1->0<br />
"<akó>"<br />
"akó" prn p1 sg @nsubj #2->1<br />
"<sa>"<br />
"sa" det def @det #3->4<br />
"<haróng>"<br />
"haróng" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<dakúla>"<br />
"dakúla" adj sg @amod #6->4<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
* Translating to "I left the big house", this sentence demonstrates the most common structure of adjectival modifiers in Bikol, where the noun is followed by a linking word and the adjective. Dakúla (big), has an amod relation to the noun haróng (house).<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Although syntactically this sentence is very similar to the previous example, with a noun followed by a linking word and an adjective, this demonstrates a different kind of adjectival modifier. The adjective "nakadukót" is derived from the verb stem "dukót", meaning to glue or attach. The "naka" prefix changes the word into an adjective, meaning attached or stuck on. The sentence translates to, "Remove the poster attached to the bridge."<br />
<br />
=== obj ===<br />
The object relation relates a verb and its object. Bikol is a VSO language, so the object of the verb is usually found after the subject, like in the following two examples.<br />
<br />
<pre><br />
"<Iinulí>"<br />
"ulí" vblex pres @root #1->0<br />
"<ko>"<br />
"ko" prn p1 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<libró>"<br />
"libró" n @obj #4->1<br />
"<.>"<br />
"." sent @punct #5->1<br />
</pre><br />
* In this sentence, which translates to "I return a book", libró is the object and depends on the verb, iinulí.<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Similarly, pósters is the object of this sentence and depends on the verb, ipaháli.<br />
<br />
=== advmod ===<br />
An adverbial modifier of a word is an adverb or adverbial phrase that modifies a predicate or modifier word.<br />
<br />
<pre><br />
"<Patosón>"<br />
"patós" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pansít>"<br />
"pansít" n @obj #4->1<br />
"<,>"<br />
"," cm @punct #5->6<br />
"<tápos>"<br />
"tápos" cnjadv @conj #6->1<br />
"<,>"<br />
"," cm @punct #7->8<br />
"<darahón>"<br />
"dará" vblex inf @conj #8->1<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #9->8<br />
"<digdí>"<br />
"digdí" adv @advmod #10->8<br />
"<.>"<br />
"." sent @punct #11->1<br />
</pre><br />
* In this sentence, digdí (or here) is the adverbial modifier that modifies the verb darahón (the command form of 'bring'). <br />
<br />
<pre><br />
"<Tápos>"<br />
"tápos" adj sg @root #1->0<br />
"<na>"<br />
"na" adv @advmod #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<prográma>"<br />
"prográma" n @nsubj #4->1<br />
"<sa>"<br />
"sa" pr @case #5->6<br />
"<gym>"<br />
"gym" n @nmod #6->4<br />
"<?>"<br />
"?" sent @punct #7->1<br />
</pre><br />
*This sentence translates to "Is the program in the gym already done?" The predicate is an implied "is", so the root the adjective, tápos (done). The adverbial modifier na (already) describes this adjective.<br />
<br />
<br />
[[Category: sp17_UD]]<br />
[[Category: Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=Bikol/Universal_Dependencies&diff=4981Bikol/Universal Dependencies2017-05-01T22:39:21Z<p>Eortici1: </p>
<hr />
<div>This page documents Bikol Universal Dependencies. <br />
<br />
== Evaluation ==<br />
<br />
This assignment involved two Bikol corpora. The first corpus, and the larger of the two, was used to train two parsers (withmorph and nomorph). Then, each of the two parsers were evaluated using both corpora, Corpus 1 and a smaller Corpus 2.<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|<br />
|Sentences<br />
|Forms<br />
|-<br />
|Corpus 1<br />
|32<br />
|198<br />
|-<br />
|Corpus 2<br />
|16<br />
|111<br />
|}<br />
<br />
The following are the LAS (Label Attachment Score) and UAS (Unlabeled Attachment Score) results when evaluating the parser trained with morphological information (bcl.withmorph.udpipe) using Corpus 1 and Corpus 2.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|[C1 LAS]<br />
|[C2 UAS]<br />
|-<br />
|Corpus 2<br />
|[C2 LAS]<br />
|[C2 UAS]<br />
|}<br />
<br />
The following are the LAS and UAS results when evaluating the parser trained with morphological information removed (bcl.nomorph.udpipe) using Corpus 1 and Corpus 2.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|[C1 LAS]<br />
|[C2 UAS]<br />
|-<br />
|Corpus 2<br />
|[C2 LAS]<br />
|[C2 UAS]<br />
|}<br />
<br />
== Dependency relations ==<br />
The following five sections illustrate examples of dependency relations used frequently in Bikol.<br />
<br />
=== advcl ===<br />
According to the Universal Dependencies documentation, an adverbial clause modifier is a clause which modifies a verb or other predicate as a modifier and not as a core complement. In my corpora, this relation came up in if statements.<br />
<br />
<pre><br />
"<Kon>"<br />
"kon" cnjsub @advmod #1->2<br />
"<makuá>"<br />
"kuá" vblex inf @advcl #2->5<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #3->2<br />
"<,>"<br />
"," cm @punct #4->5<br />
"<itaó>"<br />
"taó" vblex inf @root #5->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #6->5<br />
"<saná>"<br />
"saná" adv @advmod #7->5<br />
"<sakó>"<br />
"sakó" prn p1 sg @iobj #8->5<br />
"<.>"<br />
"." sent @punct #9->5<br />
</pre><br />
* This sentence translates it to "If you find it, just give it to me." The first clause, "If you find it", is an adverbial clause modifier which modifies "just give it to me". Thus, the root of the adverbial clause modifier, makuá, has an advcl dependency on the root of the independent clause that it modifies, itaó.<br />
<br />
<pre><br />
"<Kon>"<br />
"Kon" cnjsub @advmod #1->2<br />
"<maglúto>"<br />
"lúto" vblex inf @advcl #2->7<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #3->2<br />
"<nin>"<br />
"nin" link @clf #4->5<br />
"<jéllo>"<br />
"jéllo" n @obj #5->2<br />
"<,>"<br />
"," cm @punct #6->7<br />
"<magkuá>"<br />
"kuá" vblex inf @root #7->0<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #8->7<br />
"<nin>"<br />
"nin" link @clf #9->10<br />
"<pláto>"<br />
"pláto" n @obj #10->7<br />
"<sagkód>"<br />
"sagkód" adj sg @cc #11->12 ! and<br />
"<kutsára>"<br />
"kutsára" n @conj #12->10<br />
"<.>"<br />
"." sent @punct #13->2<br />
</pre><br />
* This sentence essentially has the same structure as the previous example. The root of the adverbial clause modifier, maglúto, has an advcl dependency on magkuá, the root of the independent clause.<br />
<br />
=== obl ===<br />
The obl (oblique) relation is used for a nominal that serves as a non-core argument or adjunct. In Bikol, this typically means that it functions as an adverbial attached to a predicate.<br />
<br />
<pre><br />
"<Hanápon>"<br />
"hánap" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<siyá>"<br />
"siyá" prn p3 sg @obj #3->1<br />
"<sa>"<br />
"sa" pr @case #4->5<br />
"<cafeteria>"<br />
"cafeteria" n @obl #5->1<br />
"<.>"<br />
"." sent @punct #6->1<br />
</pre><br />
* In this sentence, meaning "Look for him/her <i>in the cafeteria</i>", cafeteria is a nominal serving as a non-core argument and modifying the verb, hanápon (to look). Therefore cafeteria has an obl relation to hanápon.<br />
<br />
<pre><br />
"<Sublión>"<br />
"sublí" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<si>"<br />
"si" det def @det #3->4<br />
"<bóla>"<br />
"bóla" n @obj #4->1<br />
"<ki>"<br />
"ki" pr @case #5->6<br />
"<Jacób>"<br />
"Jacób" cog @obl #6->1<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
*This sentence behaves similarly as the previous example, except now the nominal in the oblique relation is a proper name (Jacób). It means "Borrow the ball from Jacob".<br />
<br />
=== amod ===<br />
The amod relation refers to adjectival modifier, which simply modifies the meaning of a noun.<br />
<br />
<pre><br />
"<Nagdumán>"<br />
"dumán" vblex past @root #1->0<br />
"<akó>"<br />
"akó" prn p1 sg @nsubj #2->1<br />
"<sa>"<br />
"sa" det def @det #3->4<br />
"<haróng>"<br />
"haróng" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<dakúla>"<br />
"dakúla" adj sg @amod #6->4<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
* Translating to "I left the big house", this sentence demonstrates the most common structure of adjectival modifiers in Bikol, where the noun is followed by a linking word and the adjective. Dakúla (big), has an amod relation to the noun haróng (house).<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Although syntactically this sentence is very similar to the previous example, with a noun followed by a linking word and an adjective, this demonstrates a different kind of adjectival modifier. The adjective "nakadukót" is derived from the verb stem "dukót", meaning to glue or attach. The "naka" prefix changes the word into an adjective, meaning attached or stuck on. The sentence translates to, "Remove the poster attached to the bridge."<br />
<br />
=== obj ===<br />
The object relation relates a verb and its object. Bikol is a VSO language, so the object of the verb is usually found after the subject, like in the following two examples.<br />
<br />
<pre><br />
"<Iinulí>"<br />
"ulí" vblex pres @root #1->0<br />
"<ko>"<br />
"ko" prn p1 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<libró>"<br />
"libró" n @obj #4->1<br />
"<.>"<br />
"." sent @punct #5->1<br />
</pre><br />
* In this sentence, which translates to "I return a book", libró is the object and depends on the verb, iinulí.<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Similarly, pósters is the object of this sentence and depends on the verb, ipaháli.<br />
<br />
=== advmod ===<br />
An adverbial modifier of a word is an adverb or adverbial phrase that modifies a predicate or modifier word.<br />
<br />
<pre><br />
"<Patosón>"<br />
"patós" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pansít>"<br />
"pansít" n @obj #4->1<br />
"<,>"<br />
"," cm @punct #5->6<br />
"<tápos>"<br />
"tápos" cnjadv @conj #6->1<br />
"<,>"<br />
"," cm @punct #7->8<br />
"<darahón>"<br />
"dará" vblex inf @conj #8->1<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #9->8<br />
"<digdí>"<br />
"digdí" adv @advmod #10->8<br />
"<.>"<br />
"." sent @punct #11->1<br />
</pre><br />
* In this sentence, digdí (or here) is the adverbial modifier that modifies the verb darahón (the command form of 'bring'). <br />
<br />
<pre><br />
"<Tápos>"<br />
"tápos" adj sg @root #1->0<br />
"<na>"<br />
"na" adv @advmod #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<prográma>"<br />
"prográma" n @nsubj #4->1<br />
"<sa>"<br />
"sa" pr @case #5->6<br />
"<gym>"<br />
"gym" n @nmod #6->4<br />
"<?>"<br />
"?" sent @punct #7->1<br />
</pre><br />
*This sentence translates to "Is the program in the gym already done?" The predicate is an implied "is", so the root the adjective, tápos (done). The adverbial modifier na (already) describes this adjective.<br />
<br />
<br />
[[Category: sp17_UD]]<br />
[[Category: Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=Bikol/Universal_Dependencies&diff=4980Bikol/Universal Dependencies2017-05-01T22:35:38Z<p>Eortici1: </p>
<hr />
<div>This page documents Bikol Universal Dependencies. <br />
<br />
== Evaluation ==<br />
<br />
This assignment involved two Bikol corpora. The first corpus, and the larger of the two, was used to train two parsers (withmorph and nomorph). Then, each of the two parsers were evaluated using both corpora, Corpus 1 and a smaller Corpus 2.<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|<br />
|Sentences<br />
|Forms<br />
|-<br />
|Corpus 1<br />
|C1 32<br />
|C1 198<br />
|-<br />
|Corpus 2<br />
|C2 sent<br />
|C2 form<br />
|}<br />
<br />
The following are the LAS (Label Attachment Score) and UAS (Unlabeled Attachment Score) results when evaluating the parser trained with morphological information (bcl.withmorph.udpipe) using Corpus 1 and Corpus 2.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|[C1 LAS]<br />
|[C2 UAS]<br />
|-<br />
|Corpus 2<br />
|[C2 LAS]<br />
|[C2 UAS]<br />
|}<br />
<br />
The following are the LAS and UAS results when evaluating the parser trained with morphological information removed (bcl.nomorph.udpipe) using Corpus 1 and Corpus 2.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|[C1 LAS]<br />
|[C2 UAS]<br />
|-<br />
|Corpus 2<br />
|[C2 LAS]<br />
|[C2 UAS]<br />
|}<br />
<br />
== Dependency relations ==<br />
The following five sections illustrate examples of dependency relations used frequently in Bikol.<br />
<br />
=== advcl ===<br />
According to the Universal Dependencies documentation, an adverbial clause modifier is a clause which modifies a verb or other predicate as a modifier and not as a core complement. In my corpora, this relation came up in if statements.<br />
<br />
<pre><br />
"<Kon>"<br />
"kon" cnjsub @advmod #1->2<br />
"<makuá>"<br />
"kuá" vblex inf @advcl #2->5<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #3->2<br />
"<,>"<br />
"," cm @punct #4->5<br />
"<itaó>"<br />
"taó" vblex inf @root #5->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #6->5<br />
"<saná>"<br />
"saná" adv @advmod #7->5<br />
"<sakó>"<br />
"sakó" prn p1 sg @iobj #8->5<br />
"<.>"<br />
"." sent @punct #9->5<br />
</pre><br />
* This sentence translates it to "If you find it, just give it to me." The first clause, "If you find it", is an adverbial clause modifier which modifies "just give it to me". Thus, the root of the adverbial clause modifier, makuá, has an advcl dependency on the root of the independent clause that it modifies, itaó.<br />
<br />
<pre><br />
"<Kon>"<br />
"Kon" cnjsub @advmod #1->2<br />
"<maglúto>"<br />
"lúto" vblex inf @advcl #2->7<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #3->2<br />
"<nin>"<br />
"nin" link @clf #4->5<br />
"<jéllo>"<br />
"jéllo" n @obj #5->2<br />
"<,>"<br />
"," cm @punct #6->7<br />
"<magkuá>"<br />
"kuá" vblex inf @root #7->0<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #8->7<br />
"<nin>"<br />
"nin" link @clf #9->10<br />
"<pláto>"<br />
"pláto" n @obj #10->7<br />
"<sagkód>"<br />
"sagkód" adj sg @cc #11->12 ! and<br />
"<kutsára>"<br />
"kutsára" n @conj #12->10<br />
"<.>"<br />
"." sent @punct #13->2<br />
</pre><br />
* This sentence essentially has the same structure as the previous example. The root of the adverbial clause modifier, maglúto, has an advcl dependency on magkuá, the root of the independent clause.<br />
<br />
=== obl ===<br />
The obl (oblique) relation is used for a nominal that serves as a non-core argument or adjunct. In Bikol, this typically means that it functions as an adverbial attached to a predicate.<br />
<br />
<pre><br />
"<Hanápon>"<br />
"hánap" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<siyá>"<br />
"siyá" prn p3 sg @obj #3->1<br />
"<sa>"<br />
"sa" pr @case #4->5<br />
"<cafeteria>"<br />
"cafeteria" n @obl #5->1<br />
"<.>"<br />
"." sent @punct #6->1<br />
</pre><br />
* In this sentence, meaning "Look for him/her <i>in the cafeteria</i>", cafeteria is a nominal serving as a non-core argument and modifying the verb, hanápon (to look). Therefore cafeteria has an obl relation to hanápon.<br />
<br />
<pre><br />
"<Sublión>"<br />
"sublí" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<si>"<br />
"si" det def @det #3->4<br />
"<bóla>"<br />
"bóla" n @obj #4->1<br />
"<ki>"<br />
"ki" pr @case #5->6<br />
"<Jacób>"<br />
"Jacób" cog @obl #6->1<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
*This sentence behaves similarly as the previous example, except now the nominal in the oblique relation is a proper name (Jacób). It means "Borrow the ball from Jacob".<br />
<br />
=== amod ===<br />
The amod relation refers to adjectival modifier, which simply modifies the meaning of a noun.<br />
<br />
<pre><br />
"<Nagdumán>"<br />
"dumán" vblex past @root #1->0<br />
"<akó>"<br />
"akó" prn p1 sg @nsubj #2->1<br />
"<sa>"<br />
"sa" det def @det #3->4<br />
"<haróng>"<br />
"haróng" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<dakúla>"<br />
"dakúla" adj sg @amod #6->4<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
* Translating to "I left the big house", this sentence demonstrates the most common structure of adjectival modifiers in Bikol, where the noun is followed by a linking word and the adjective. Dakúla (big), has an amod relation to the noun haróng (house).<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Although syntactically this sentence is very similar to the previous example, with a noun followed by a linking word and an adjective, this demonstrates a different kind of adjectival modifier. The adjective "nakadukót" is derived from the verb stem "dukót", meaning to glue or attach. The "naka" prefix changes the word into an adjective, meaning attached or stuck on. The sentence translates to, "Remove the poster attached to the bridge."<br />
<br />
=== obj ===<br />
The object relation relates a verb and its object. Bikol is a VSO language, so the object of the verb is usually found after the subject, like in the following two examples.<br />
<br />
<pre><br />
"<Iinulí>"<br />
"ulí" vblex pres @root #1->0<br />
"<ko>"<br />
"ko" prn p1 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<libró>"<br />
"libró" n @obj #4->1<br />
"<.>"<br />
"." sent @punct #5->1<br />
</pre><br />
* In this sentence, which translates to "I return a book", libró is the object and depends on the verb, iinulí.<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Similarly, pósters is the object of this sentence and depends on the verb, ipaháli.<br />
<br />
=== advmod ===<br />
An adverbial modifier of a word is an adverb or adverbial phrase that modifies a predicate or modifier word.<br />
<br />
<pre><br />
"<Patosón>"<br />
"patós" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pansít>"<br />
"pansít" n @obj #4->1<br />
"<,>"<br />
"," cm @punct #5->6<br />
"<tápos>"<br />
"tápos" cnjadv @conj #6->1<br />
"<,>"<br />
"," cm @punct #7->8<br />
"<darahón>"<br />
"dará" vblex inf @conj #8->1<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #9->8<br />
"<digdí>"<br />
"digdí" adv @advmod #10->8<br />
"<.>"<br />
"." sent @punct #11->1<br />
</pre><br />
* In this sentence, digdí (or here) is the adverbial modifier that modifies the verb darahón (the command form of 'bring'). <br />
<br />
<pre><br />
"<Tápos>"<br />
"tápos" adj sg @root #1->0<br />
"<na>"<br />
"na" adv @advmod #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<prográma>"<br />
"prográma" n @nsubj #4->1<br />
"<sa>"<br />
"sa" pr @case #5->6<br />
"<gym>"<br />
"gym" n @nmod #6->4<br />
"<?>"<br />
"?" sent @punct #7->1<br />
</pre><br />
*This sentence translates to "Is the program in the gym already done?" The predicate is an implied "is", so the root the adjective, tápos (done). The adverbial modifier na (already) describes this adjective.<br />
<br />
<br />
[[Category: sp17_UD]]<br />
[[Category: Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=Bikol/Universal_Dependencies&diff=4913Bikol/Universal Dependencies2017-04-26T19:25:02Z<p>Eortici1: </p>
<hr />
<div>This page documents Bikol Universal Dependencies. <br />
<br />
== Evaluation ==<br />
<br />
This assignment involved two Bikol corpora. The first corpus, and the larger of the two, was used to train two parsers (withmorph and nomorph). Then, each of the two parsers were evaluated using both corpora, Corpus 1 and a smaller Corpus 2.<br />
<br />
{|class="wikitable"<br />
|+Corpus Size<br />
|-<br />
|<br />
|Sentences<br />
|Forms<br />
|-<br />
|Corpus 1<br />
|C1 sent<br />
|C1 form<br />
|-<br />
|Corpus 2<br />
|C2 sent<br />
|C2 form<br />
|}<br />
<br />
The following are the LAS (Label Attachment Score) and UAS (Unlabeled Attachment Score) results when evaluating the parser trained with morphological information (bcl.withmorph.udpipe) using Corpus 1 and Corpus 2.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.withmorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|[C1 LAS]<br />
|[C2 UAS]<br />
|-<br />
|Corpus 2<br />
|[C2 LAS]<br />
|[C2 UAS]<br />
|}<br />
<br />
The following are the LAS and UAS results when evaluating the parser trained with morphological information removed (bcl.nomorph.udpipe) using Corpus 1 and Corpus 2.<br />
<br />
{| class="wikitable"<br />
|+Evaluation of bcl.nomorph.udpipe<br />
|-<br />
|<br />
|LAS<br />
|UAS<br />
|-<br />
|Corpus 1<br />
|[C1 LAS]<br />
|[C2 UAS]<br />
|-<br />
|Corpus 2<br />
|[C2 LAS]<br />
|[C2 UAS]<br />
|}<br />
<br />
== Dependency relations ==<br />
The following five sections illustrate examples of dependency relations used frequently in Bikol.<br />
<br />
=== advcl ===<br />
According to the Universal Dependencies documentation, an adverbial clause modifier is a clause which modifies a verb or other predicate as a modifier and not as a core complement. In my corpora, this relation came up in if statements.<br />
<br />
<pre><br />
"<Kon>"<br />
"kon" cnjsub @advmod #1->2<br />
"<makuá>"<br />
"kuá" vblex inf @advcl #2->5<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #3->2<br />
"<,>"<br />
"," cm @punct #4->5<br />
"<itaó>"<br />
"taó" vblex inf @root #5->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #6->5<br />
"<saná>"<br />
"saná" adv @advmod #7->5<br />
"<sakó>"<br />
"sakó" prn p1 sg @iobj #8->5<br />
"<.>"<br />
"." sent @punct #9->5<br />
</pre><br />
* This sentence translates it to "If you find it, just give it to me." The first clause, "If you find it", is an adverbial clause modifier which modifies "just give it to me". Thus, the root of the adverbial clause modifier, makuá, has an advcl dependency on the root of the independent clause that it modifies, itaó.<br />
<br />
<pre><br />
"<Kon>"<br />
"Kon" cnjsub @advmod #1->2<br />
"<maglúto>"<br />
"lúto" vblex inf @advcl #2->7<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #3->2<br />
"<nin>"<br />
"nin" link @clf #4->5<br />
"<jéllo>"<br />
"jéllo" n @obj #5->2<br />
"<,>"<br />
"," cm @punct #6->7<br />
"<magkuá>"<br />
"kuá" vblex inf @root #7->0<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #8->7<br />
"<nin>"<br />
"nin" link @clf #9->10<br />
"<pláto>"<br />
"pláto" n @obj #10->7<br />
"<sagkód>"<br />
"sagkód" adj sg @cc #11->12 ! and<br />
"<kutsára>"<br />
"kutsára" n @conj #12->10<br />
"<.>"<br />
"." sent @punct #13->2<br />
</pre><br />
* This sentence essentially has the same structure as the previous example. The root of the adverbial clause modifier, maglúto, has an advcl dependency on magkuá, the root of the independent clause.<br />
<br />
=== obl ===<br />
The obl (oblique) relation is used for a nominal that serves as a non-core argument or adjunct. In Bikol, this typically means that it functions as an adverbial attached to a predicate.<br />
<br />
<pre><br />
"<Hanápon>"<br />
"hánap" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<siyá>"<br />
"siyá" prn p3 sg @obj #3->1<br />
"<sa>"<br />
"sa" pr @case #4->5<br />
"<cafeteria>"<br />
"cafeteria" n @obl #5->1<br />
"<.>"<br />
"." sent @punct #6->1<br />
</pre><br />
* In this sentence, meaning "Look for him/her <i>in the cafeteria</i>", cafeteria is a nominal serving as a non-core argument and modifying the verb, hanápon (to look). Therefore cafeteria has an obl relation to hanápon.<br />
<br />
<pre><br />
"<Sublión>"<br />
"sublí" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<si>"<br />
"si" det def @det #3->4<br />
"<bóla>"<br />
"bóla" n @obj #4->1<br />
"<ki>"<br />
"ki" pr @case #5->6<br />
"<Jacób>"<br />
"Jacób" cog @obl #6->1<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
*This sentence behaves similarly as the previous example, except now the nominal in the oblique relation is a proper name (Jacób). It means "Borrow the ball from Jacob".<br />
<br />
=== amod ===<br />
The amod relation refers to adjectival modifier, which simply modifies the meaning of a noun.<br />
<br />
<pre><br />
"<Nagdumán>"<br />
"dumán" vblex past @root #1->0<br />
"<akó>"<br />
"akó" prn p1 sg @nsubj #2->1<br />
"<sa>"<br />
"sa" det def @det #3->4<br />
"<haróng>"<br />
"haróng" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<dakúla>"<br />
"dakúla" adj sg @amod #6->4<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
* Translating to "I left the big house", this sentence demonstrates the most common structure of adjectival modifiers in Bikol, where the noun is followed by a linking word and the adjective. Dakúla (big), has an amod relation to the noun haróng (house).<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Although syntactically this sentence is very similar to the previous example, with a noun followed by a linking word and an adjective, this demonstrates a different kind of adjectival modifier. The adjective "nakadukót" is derived from the verb stem "dukót", meaning to glue or attach. The "naka" prefix changes the word into an adjective, meaning attached or stuck on. The sentence translates to, "Remove the poster attached to the bridge."<br />
<br />
=== obj ===<br />
The object relation relates a verb and its object. Bikol is a VSO language, so the object of the verb is usually found after the subject, like in the following two examples.<br />
<br />
<pre><br />
"<Iinulí>"<br />
"ulí" vblex pres @root #1->0<br />
"<ko>"<br />
"ko" prn p1 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<libró>"<br />
"libró" n @obj #4->1<br />
"<.>"<br />
"." sent @punct #5->1<br />
</pre><br />
* In this sentence, which translates to "I return a book", libró is the object and depends on the verb, iinulí.<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Similarly, pósters is the object of this sentence and depends on the verb, ipaháli.<br />
<br />
=== advmod ===<br />
An adverbial modifier of a word is an adverb or adverbial phrase that modifies a predicate or modifier word.<br />
<br />
<pre><br />
"<Patosón>"<br />
"patós" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pansít>"<br />
"pansít" n @obj #4->1<br />
"<,>"<br />
"," cm @punct #5->6<br />
"<tápos>"<br />
"tápos" cnjadv @conj #6->1<br />
"<,>"<br />
"," cm @punct #7->8<br />
"<darahón>"<br />
"dará" vblex inf @conj #8->1<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #9->8<br />
"<digdí>"<br />
"digdí" adv @advmod #10->8<br />
"<.>"<br />
"." sent @punct #11->1<br />
</pre><br />
* In this sentence, digdí (or here) is the adverbial modifier that modifies the verb darahón (the command form of 'bring'). <br />
<br />
<pre><br />
"<Tápos>"<br />
"tápos" adj sg @root #1->0<br />
"<na>"<br />
"na" adv @advmod #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<prográma>"<br />
"prográma" n @nsubj #4->1<br />
"<sa>"<br />
"sa" pr @case #5->6<br />
"<gym>"<br />
"gym" n @nmod #6->4<br />
"<?>"<br />
"?" sent @punct #7->1<br />
</pre><br />
*This sentence translates to "Is the program in the gym already done?" The predicate is an implied "is", so the root the adjective, tápos (done). The adverbial modifier na (already) describes this adjective.<br />
<br />
<br />
[[Category: sp17_UD]]<br />
[[Category: Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=Bikol/Universal_Dependencies&diff=4912Bikol/Universal Dependencies2017-04-26T19:13:47Z<p>Eortici1: /* Dependency relations */</p>
<hr />
<div>This page documents Bikol Universal Dependencies. <br />
<br />
== Evaluation ==<br />
<br />
== Dependency relations ==<br />
The following five sections illustrate examples of dependency relations used frequently in Bikol.<br />
<br />
=== advcl ===<br />
According to the Universal Dependencies documentation, an adverbial clause modifier is a clause which modifies a verb or other predicate as a modifier and not as a core complement. In my corpora, this relation came up in if statements.<br />
<br />
<pre><br />
"<Kon>"<br />
"kon" cnjsub @advmod #1->2<br />
"<makuá>"<br />
"kuá" vblex inf @advcl #2->5<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #3->2<br />
"<,>"<br />
"," cm @punct #4->5<br />
"<itaó>"<br />
"taó" vblex inf @root #5->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #6->5<br />
"<saná>"<br />
"saná" adv @advmod #7->5<br />
"<sakó>"<br />
"sakó" prn p1 sg @iobj #8->5<br />
"<.>"<br />
"." sent @punct #9->5<br />
</pre><br />
* This sentence translates it to "If you find it, just give it to me." The first clause, "If you find it", is an adverbial clause modifier which modifies "just give it to me". Thus, the root of the adverbial clause modifier, makuá, has an advcl dependency on the root of the independent clause that it modifies, itaó.<br />
<br />
<pre><br />
"<Kon>"<br />
"Kon" cnjsub @advmod #1->2<br />
"<maglúto>"<br />
"lúto" vblex inf @advcl #2->7<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #3->2<br />
"<nin>"<br />
"nin" link @clf #4->5<br />
"<jéllo>"<br />
"jéllo" n @obj #5->2<br />
"<,>"<br />
"," cm @punct #6->7<br />
"<magkuá>"<br />
"kuá" vblex inf @root #7->0<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #8->7<br />
"<nin>"<br />
"nin" link @clf #9->10<br />
"<pláto>"<br />
"pláto" n @obj #10->7<br />
"<sagkód>"<br />
"sagkód" adj sg @cc #11->12 ! and<br />
"<kutsára>"<br />
"kutsára" n @conj #12->10<br />
"<.>"<br />
"." sent @punct #13->2<br />
</pre><br />
* This sentence essentially has the same structure as the previous example. The root of the adverbial clause modifier, maglúto, has an advcl dependency on magkuá, the root of the independent clause.<br />
<br />
=== obl ===<br />
The obl (oblique) relation is used for a nominal that serves as a non-core argument or adjunct. In Bikol, this typically means that it functions as an adverbial attached to a predicate.<br />
<br />
<pre><br />
"<Hanápon>"<br />
"hánap" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<siyá>"<br />
"siyá" prn p3 sg @obj #3->1<br />
"<sa>"<br />
"sa" pr @case #4->5<br />
"<cafeteria>"<br />
"cafeteria" n @obl #5->1<br />
"<.>"<br />
"." sent @punct #6->1<br />
</pre><br />
* In this sentence, meaning "Look for him/her <i>in the cafeteria</i>", cafeteria is a nominal serving as a non-core argument and modifying the verb, hanápon (to look). Therefore cafeteria has an obl relation to hanápon.<br />
<br />
<pre><br />
"<Sublión>"<br />
"sublí" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<si>"<br />
"si" det def @det #3->4<br />
"<bóla>"<br />
"bóla" n @obj #4->1<br />
"<ki>"<br />
"ki" pr @case #5->6<br />
"<Jacób>"<br />
"Jacób" cog @obl #6->1<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
*This sentence behaves similarly as the previous example, except now the nominal in the oblique relation is a proper name (Jacób). It means "Borrow the ball from Jacob".<br />
<br />
=== amod ===<br />
The amod relation refers to adjectival modifier, which simply modifies the meaning of a noun.<br />
<br />
<pre><br />
"<Nagdumán>"<br />
"dumán" vblex past @root #1->0<br />
"<akó>"<br />
"akó" prn p1 sg @nsubj #2->1<br />
"<sa>"<br />
"sa" det def @det #3->4<br />
"<haróng>"<br />
"haróng" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<dakúla>"<br />
"dakúla" adj sg @amod #6->4<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
* Translating to "I left the big house", this sentence demonstrates the most common structure of adjectival modifiers in Bikol, where the noun is followed by a linking word and the adjective. Dakúla (big), has an amod relation to the noun haróng (house).<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Although syntactically this sentence is very similar to the previous example, with a noun followed by a linking word and an adjective, this demonstrates a different kind of adjectival modifier. The adjective "nakadukót" is derived from the verb stem "dukót", meaning to glue or attach. The "naka" prefix changes the word into an adjective, meaning attached or stuck on. The sentence translates to, "Remove the poster attached to the bridge."<br />
<br />
=== obj ===<br />
The object relation relates a verb and its object. Bikol is a VSO language, so the object of the verb is usually found after the subject, like in the following two examples.<br />
<br />
<pre><br />
"<Iinulí>"<br />
"ulí" vblex pres @root #1->0<br />
"<ko>"<br />
"ko" prn p1 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<libró>"<br />
"libró" n @obj #4->1<br />
"<.>"<br />
"." sent @punct #5->1<br />
</pre><br />
* In this sentence, which translates to "I return a book", libró is the object and depends on the verb, iinulí.<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Similarly, pósters is the object of this sentence and depends on the verb, ipaháli.<br />
<br />
=== advmod ===<br />
An adverbial modifier of a word is an adverb or adverbial phrase that modifies a predicate or modifier word.<br />
<br />
<pre><br />
"<Patosón>"<br />
"patós" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pansít>"<br />
"pansít" n @obj #4->1<br />
"<,>"<br />
"," cm @punct #5->6<br />
"<tápos>"<br />
"tápos" cnjadv @conj #6->1<br />
"<,>"<br />
"," cm @punct #7->8<br />
"<darahón>"<br />
"dará" vblex inf @conj #8->1<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #9->8<br />
"<digdí>"<br />
"digdí" adv @advmod #10->8<br />
"<.>"<br />
"." sent @punct #11->1<br />
</pre><br />
* In this sentence, digdí (or here) is the adverbial modifier that modifies the verb darahón (the command form of 'bring'). <br />
<br />
<pre><br />
"<Tápos>"<br />
"tápos" adj sg @root #1->0<br />
"<na>"<br />
"na" adv @advmod #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<prográma>"<br />
"prográma" n @nsubj #4->1<br />
"<sa>"<br />
"sa" pr @case #5->6<br />
"<gym>"<br />
"gym" n @nmod #6->4<br />
"<?>"<br />
"?" sent @punct #7->1<br />
</pre><br />
*This sentence translates to "Is the program in the gym already done?" The predicate is an implied "is", so the root the adjective, tápos (done). The adverbial modifier na (already) describes this adjective.<br />
<br />
<br />
[[Category: sp17_UD]]<br />
[[Category: Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=Bikol/Universal_Dependencies&diff=4911Bikol/Universal Dependencies2017-04-26T19:13:39Z<p>Eortici1: /* Dependency relations */</p>
<hr />
<div>This page documents Bikol Universal Dependencies. <br />
<br />
== Evaluation ==<br />
<br />
== Dependency relations ==<br />
The following five sections illustrate examples of 5 dependency relations used frequently in Bikol.<br />
<br />
=== advcl ===<br />
According to the Universal Dependencies documentation, an adverbial clause modifier is a clause which modifies a verb or other predicate as a modifier and not as a core complement. In my corpora, this relation came up in if statements.<br />
<br />
<pre><br />
"<Kon>"<br />
"kon" cnjsub @advmod #1->2<br />
"<makuá>"<br />
"kuá" vblex inf @advcl #2->5<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #3->2<br />
"<,>"<br />
"," cm @punct #4->5<br />
"<itaó>"<br />
"taó" vblex inf @root #5->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #6->5<br />
"<saná>"<br />
"saná" adv @advmod #7->5<br />
"<sakó>"<br />
"sakó" prn p1 sg @iobj #8->5<br />
"<.>"<br />
"." sent @punct #9->5<br />
</pre><br />
* This sentence translates it to "If you find it, just give it to me." The first clause, "If you find it", is an adverbial clause modifier which modifies "just give it to me". Thus, the root of the adverbial clause modifier, makuá, has an advcl dependency on the root of the independent clause that it modifies, itaó.<br />
<br />
<pre><br />
"<Kon>"<br />
"Kon" cnjsub @advmod #1->2<br />
"<maglúto>"<br />
"lúto" vblex inf @advcl #2->7<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #3->2<br />
"<nin>"<br />
"nin" link @clf #4->5<br />
"<jéllo>"<br />
"jéllo" n @obj #5->2<br />
"<,>"<br />
"," cm @punct #6->7<br />
"<magkuá>"<br />
"kuá" vblex inf @root #7->0<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #8->7<br />
"<nin>"<br />
"nin" link @clf #9->10<br />
"<pláto>"<br />
"pláto" n @obj #10->7<br />
"<sagkód>"<br />
"sagkód" adj sg @cc #11->12 ! and<br />
"<kutsára>"<br />
"kutsára" n @conj #12->10<br />
"<.>"<br />
"." sent @punct #13->2<br />
</pre><br />
* This sentence essentially has the same structure as the previous example. The root of the adverbial clause modifier, maglúto, has an advcl dependency on magkuá, the root of the independent clause.<br />
<br />
=== obl ===<br />
The obl (oblique) relation is used for a nominal that serves as a non-core argument or adjunct. In Bikol, this typically means that it functions as an adverbial attached to a predicate.<br />
<br />
<pre><br />
"<Hanápon>"<br />
"hánap" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<siyá>"<br />
"siyá" prn p3 sg @obj #3->1<br />
"<sa>"<br />
"sa" pr @case #4->5<br />
"<cafeteria>"<br />
"cafeteria" n @obl #5->1<br />
"<.>"<br />
"." sent @punct #6->1<br />
</pre><br />
* In this sentence, meaning "Look for him/her <i>in the cafeteria</i>", cafeteria is a nominal serving as a non-core argument and modifying the verb, hanápon (to look). Therefore cafeteria has an obl relation to hanápon.<br />
<br />
<pre><br />
"<Sublión>"<br />
"sublí" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<si>"<br />
"si" det def @det #3->4<br />
"<bóla>"<br />
"bóla" n @obj #4->1<br />
"<ki>"<br />
"ki" pr @case #5->6<br />
"<Jacób>"<br />
"Jacób" cog @obl #6->1<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
*This sentence behaves similarly as the previous example, except now the nominal in the oblique relation is a proper name (Jacób). It means "Borrow the ball from Jacob".<br />
<br />
=== amod ===<br />
The amod relation refers to adjectival modifier, which simply modifies the meaning of a noun.<br />
<br />
<pre><br />
"<Nagdumán>"<br />
"dumán" vblex past @root #1->0<br />
"<akó>"<br />
"akó" prn p1 sg @nsubj #2->1<br />
"<sa>"<br />
"sa" det def @det #3->4<br />
"<haróng>"<br />
"haróng" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<dakúla>"<br />
"dakúla" adj sg @amod #6->4<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
* Translating to "I left the big house", this sentence demonstrates the most common structure of adjectival modifiers in Bikol, where the noun is followed by a linking word and the adjective. Dakúla (big), has an amod relation to the noun haróng (house).<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Although syntactically this sentence is very similar to the previous example, with a noun followed by a linking word and an adjective, this demonstrates a different kind of adjectival modifier. The adjective "nakadukót" is derived from the verb stem "dukót", meaning to glue or attach. The "naka" prefix changes the word into an adjective, meaning attached or stuck on. The sentence translates to, "Remove the poster attached to the bridge."<br />
<br />
=== obj ===<br />
The object relation relates a verb and its object. Bikol is a VSO language, so the object of the verb is usually found after the subject, like in the following two examples.<br />
<br />
<pre><br />
"<Iinulí>"<br />
"ulí" vblex pres @root #1->0<br />
"<ko>"<br />
"ko" prn p1 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<libró>"<br />
"libró" n @obj #4->1<br />
"<.>"<br />
"." sent @punct #5->1<br />
</pre><br />
* In this sentence, which translates to "I return a book", libró is the object and depends on the verb, iinulí.<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Similarly, pósters is the object of this sentence and depends on the verb, ipaháli.<br />
<br />
=== advmod ===<br />
An adverbial modifier of a word is an adverb or adverbial phrase that modifies a predicate or modifier word.<br />
<br />
<pre><br />
"<Patosón>"<br />
"patós" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pansít>"<br />
"pansít" n @obj #4->1<br />
"<,>"<br />
"," cm @punct #5->6<br />
"<tápos>"<br />
"tápos" cnjadv @conj #6->1<br />
"<,>"<br />
"," cm @punct #7->8<br />
"<darahón>"<br />
"dará" vblex inf @conj #8->1<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #9->8<br />
"<digdí>"<br />
"digdí" adv @advmod #10->8<br />
"<.>"<br />
"." sent @punct #11->1<br />
</pre><br />
* In this sentence, digdí (or here) is the adverbial modifier that modifies the verb darahón (the command form of 'bring'). <br />
<br />
<pre><br />
"<Tápos>"<br />
"tápos" adj sg @root #1->0<br />
"<na>"<br />
"na" adv @advmod #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<prográma>"<br />
"prográma" n @nsubj #4->1<br />
"<sa>"<br />
"sa" pr @case #5->6<br />
"<gym>"<br />
"gym" n @nmod #6->4<br />
"<?>"<br />
"?" sent @punct #7->1<br />
</pre><br />
*This sentence translates to "Is the program in the gym already done?" The predicate is an implied "is", so the root the adjective, tápos (done). The adverbial modifier na (already) describes this adjective.<br />
<br />
<br />
[[Category: sp17_UD]]<br />
[[Category: Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=Bikol/Universal_Dependencies&diff=4910Bikol/Universal Dependencies2017-04-26T19:11:02Z<p>Eortici1: </p>
<hr />
<div>This page documents Bikol Universal Dependencies. <br />
<br />
== Evaluation ==<br />
<br />
== Dependency relations ==<br />
<br />
=== advcl ===<br />
According to the Universal Dependencies documentation, an adverbial clause modifier is a clause which modifies a verb or other predicate as a modifier and not as a core complement. In my corpora, this relation came up in if statements.<br />
<br />
<pre><br />
"<Kon>"<br />
"kon" cnjsub @advmod #1->2<br />
"<makuá>"<br />
"kuá" vblex inf @advcl #2->5<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #3->2<br />
"<,>"<br />
"," cm @punct #4->5<br />
"<itaó>"<br />
"taó" vblex inf @root #5->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #6->5<br />
"<saná>"<br />
"saná" adv @advmod #7->5<br />
"<sakó>"<br />
"sakó" prn p1 sg @iobj #8->5<br />
"<.>"<br />
"." sent @punct #9->5<br />
</pre><br />
* This sentence translates it to "If you find it, just give it to me." The first clause, "If you find it", is an adverbial clause modifier which modifies "just give it to me". Thus, the root of the adverbial clause modifier, makuá, has an advcl dependency on the root of the independent clause that it modifies, itaó.<br />
<br />
<pre><br />
"<Kon>"<br />
"Kon" cnjsub @advmod #1->2<br />
"<maglúto>"<br />
"lúto" vblex inf @advcl #2->7<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #3->2<br />
"<nin>"<br />
"nin" link @clf #4->5<br />
"<jéllo>"<br />
"jéllo" n @obj #5->2<br />
"<,>"<br />
"," cm @punct #6->7<br />
"<magkuá>"<br />
"kuá" vblex inf @root #7->0<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #8->7<br />
"<nin>"<br />
"nin" link @clf #9->10<br />
"<pláto>"<br />
"pláto" n @obj #10->7<br />
"<sagkód>"<br />
"sagkód" adj sg @cc #11->12 ! and<br />
"<kutsára>"<br />
"kutsára" n @conj #12->10<br />
"<.>"<br />
"." sent @punct #13->2<br />
</pre><br />
* This sentence essentially has the same structure as the previous example. The root of the adverbial clause modifier, maglúto, has an advcl dependency on magkuá, the root of the independent clause.<br />
<br />
=== obl ===<br />
The obl (oblique) relation is used for a nominal that serves as a non-core argument or adjunct. In Bikol, this typically means that it functions as an adverbial attached to a predicate.<br />
<br />
<pre><br />
"<Hanápon>"<br />
"hánap" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<siyá>"<br />
"siyá" prn p3 sg @obj #3->1<br />
"<sa>"<br />
"sa" pr @case #4->5<br />
"<cafeteria>"<br />
"cafeteria" n @obl #5->1<br />
"<.>"<br />
"." sent @punct #6->1<br />
</pre><br />
* In this sentence, meaning "Look for him/her <i>in the cafeteria</i>", cafeteria is a nominal serving as a non-core argument and modifying the verb, hanápon (to look). Therefore cafeteria has an obl relation to hanápon.<br />
<br />
<pre><br />
"<Sublión>"<br />
"sublí" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<si>"<br />
"si" det def @det #3->4<br />
"<bóla>"<br />
"bóla" n @obj #4->1<br />
"<ki>"<br />
"ki" pr @case #5->6<br />
"<Jacób>"<br />
"Jacób" cog @obl #6->1<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
*This sentence behaves similarly as the previous example, except now the nominal in the oblique relation is a proper name (Jacób). It means "Borrow the ball from Jacob".<br />
<br />
=== amod ===<br />
The amod relation refers to adjectival modifier, which simply modifies the meaning of a noun.<br />
<br />
<pre><br />
"<Nagdumán>"<br />
"dumán" vblex past @root #1->0<br />
"<akó>"<br />
"akó" prn p1 sg @nsubj #2->1<br />
"<sa>"<br />
"sa" det def @det #3->4<br />
"<haróng>"<br />
"haróng" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<dakúla>"<br />
"dakúla" adj sg @amod #6->4<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
* Translating to "I left the big house", this sentence demonstrates the most common structure of adjectival modifiers in Bikol, where the noun is followed by a linking word and the adjective. Dakúla (big), has an amod relation to the noun haróng (house).<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Although syntactically this sentence is very similar to the previous example, with a noun followed by a linking word and an adjective, this demonstrates a different kind of adjectival modifier. The adjective "nakadukót" is derived from the verb stem "dukót", meaning to glue or attach. The "naka" prefix changes the word into an adjective, meaning attached or stuck on. The sentence translates to, "Remove the poster attached to the bridge."<br />
<br />
=== obj ===<br />
The object relation relates a verb and its object. Bikol is a VSO language, so the object of the verb is usually found after the subject, like in the following two examples.<br />
<br />
<pre><br />
"<Iinulí>"<br />
"ulí" vblex pres @root #1->0<br />
"<ko>"<br />
"ko" prn p1 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<libró>"<br />
"libró" n @obj #4->1<br />
"<.>"<br />
"." sent @punct #5->1<br />
</pre><br />
* In this sentence, which translates to "I return a book", libró is the object and depends on the verb, iinulí.<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Similarly, pósters is the object of this sentence and depends on the verb, ipaháli.<br />
<br />
=== advmod ===<br />
An adverbial modifier of a word is an adverb or adverbial phrase that modifies a predicate or modifier word.<br />
<br />
<pre><br />
"<Patosón>"<br />
"patós" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pansít>"<br />
"pansít" n @obj #4->1<br />
"<,>"<br />
"," cm @punct #5->6<br />
"<tápos>"<br />
"tápos" cnjadv @conj #6->1<br />
"<,>"<br />
"," cm @punct #7->8<br />
"<darahón>"<br />
"dará" vblex inf @conj #8->1<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #9->8<br />
"<digdí>"<br />
"digdí" adv @advmod #10->8<br />
"<.>"<br />
"." sent @punct #11->1<br />
</pre><br />
* In this sentence, digdí (or here) is the adverbial modifier that modifies the verb darahón (the command form of 'bring'). <br />
<br />
<pre><br />
"<Tápos>"<br />
"tápos" adj sg @root #1->0<br />
"<na>"<br />
"na" adv @advmod #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<prográma>"<br />
"prográma" n @nsubj #4->1<br />
"<sa>"<br />
"sa" pr @case #5->6<br />
"<gym>"<br />
"gym" n @nmod #6->4<br />
"<?>"<br />
"?" sent @punct #7->1<br />
</pre><br />
*This sentence translates to "Is the program in the gym already done?" The predicate is an implied "is", so the root the adjective, tápos (done). The adverbial modifier na (already) describes this adjective.<br />
<br />
<br />
[[Category: sp17_UD]]<br />
[[Category: Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=Bikol/Universal_Dependencies&diff=4909Bikol/Universal Dependencies2017-04-26T19:09:51Z<p>Eortici1: </p>
<hr />
<div>This page documents Bikol Universal Dependencies. <br />
<br />
== Evaluation ==<br />
<br />
== Dependency relations ==<br />
<br />
=== obj ===<br />
The object relation relates a verb and its object. Bikol is a VSO language, so the object of the verb is usually found after the subject, like in the following two examples.<br />
<br />
<pre><br />
"<Iinulí>"<br />
"ulí" vblex pres @root #1->0<br />
"<ko>"<br />
"ko" prn p1 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<libró>"<br />
"libró" n @obj #4->1<br />
"<.>"<br />
"." sent @punct #5->1<br />
</pre><br />
* In this sentence, which translates to "I return a book", libró is the object and depends on the verb, iinulí.<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Similarly, pósters is the object of this sentence and depends on the verb, ipaháli.<br />
<br />
=== advmod ===<br />
An adverbial modifier of a word is an adverb or adverbial phrase that modifies a predicate or modifier word.<br />
<br />
<pre><br />
"<Patosón>"<br />
"patós" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pansít>"<br />
"pansít" n @obj #4->1<br />
"<,>"<br />
"," cm @punct #5->6<br />
"<tápos>"<br />
"tápos" cnjadv @conj #6->1<br />
"<,>"<br />
"," cm @punct #7->8<br />
"<darahón>"<br />
"dará" vblex inf @conj #8->1<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #9->8<br />
"<digdí>"<br />
"digdí" adv @advmod #10->8<br />
"<.>"<br />
"." sent @punct #11->1<br />
</pre><br />
* In this sentence, digdí (or here) is the adverbial modifier that modifies the verb darahón (the command form of 'bring'). <br />
<br />
<pre><br />
"<Tápos>"<br />
"tápos" adj sg @root #1->0<br />
"<na>"<br />
"na" adv @advmod #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<prográma>"<br />
"prográma" n @nsubj #4->1<br />
"<sa>"<br />
"sa" pr @case #5->6<br />
"<gym>"<br />
"gym" n @nmod #6->4<br />
"<?>"<br />
"?" sent @punct #7->1<br />
</pre><br />
*This sentence translates to "Is the program in the gym already done?" The predicate is an implied "is", so the root the adjective, tápos (done). The adverbial modifier na (already) describes this adjective.<br />
<br />
=== advcl ===<br />
According to the Universal Dependencies documentation, an adverbial clause modifier is a clause which modifies a verb or other predicate as a modifier and not as a core complement. In my corpora, this relation came up in if statements.<br />
<br />
<pre><br />
"<Kon>"<br />
"kon" cnjsub @advmod #1->2<br />
"<makuá>"<br />
"kuá" vblex inf @advcl #2->5<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #3->2<br />
"<,>"<br />
"," cm @punct #4->5<br />
"<itaó>"<br />
"taó" vblex inf @root #5->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #6->5<br />
"<saná>"<br />
"saná" adv @advmod #7->5<br />
"<sakó>"<br />
"sakó" prn p1 sg @iobj #8->5<br />
"<.>"<br />
"." sent @punct #9->5<br />
</pre><br />
* This sentence translates it to "If you find it, just give it to me." The first clause, "If you find it", is an adverbial clause modifier which modifies "just give it to me". Thus, the root of the adverbial clause modifier, makuá, has an advcl dependency on the root of the independent clause that it modifies, itaó.<br />
<br />
<pre><br />
"<Kon>"<br />
"Kon" cnjsub @advmod #1->2<br />
"<maglúto>"<br />
"lúto" vblex inf @advcl #2->7<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #3->2<br />
"<nin>"<br />
"nin" link @clf #4->5<br />
"<jéllo>"<br />
"jéllo" n @obj #5->2<br />
"<,>"<br />
"," cm @punct #6->7<br />
"<magkuá>"<br />
"kuá" vblex inf @root #7->0<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #8->7<br />
"<nin>"<br />
"nin" link @clf #9->10<br />
"<pláto>"<br />
"pláto" n @obj #10->7<br />
"<sagkód>"<br />
"sagkód" adj sg @cc #11->12 ! and<br />
"<kutsára>"<br />
"kutsára" n @conj #12->10<br />
"<.>"<br />
"." sent @punct #13->2<br />
</pre><br />
* This sentence essentially has the same structure as the previous example. The root of the adverbial clause modifier, maglúto, has an advcl dependency on magkuá, the root of the independent clause.<br />
<br />
=== obl ===<br />
The obl (oblique) relation is used for a nominal that serves as a non-core argument or adjunct. In Bikol, this typically means that it functions as an adverbial attached to a predicate.<br />
<br />
<pre><br />
"<Hanápon>"<br />
"hánap" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<siyá>"<br />
"siyá" prn p3 sg @obj #3->1<br />
"<sa>"<br />
"sa" pr @case #4->5<br />
"<cafeteria>"<br />
"cafeteria" n @obl #5->1<br />
"<.>"<br />
"." sent @punct #6->1<br />
</pre><br />
* In this sentence, meaning "Look for him/her <i>in the cafeteria</i>", cafeteria is a nominal serving as a non-core argument and modifying the verb, hanápon (to look). Therefore cafeteria has an obl relation to hanápon.<br />
<br />
<pre><br />
"<Sublión>"<br />
"sublí" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<si>"<br />
"si" det def @det #3->4<br />
"<bóla>"<br />
"bóla" n @obj #4->1<br />
"<ki>"<br />
"ki" pr @case #5->6<br />
"<Jacób>"<br />
"Jacób" cog @obl #6->1<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
*This sentence behaves similarly as the previous example, except now the nominal in the oblique relation is a proper name (Jacób). It means "Borrow the ball from Jacob".<br />
<br />
=== amod ===<br />
The amod relation refers to adjectival modifier, which simply modifies the meaning of a noun.<br />
<br />
<pre><br />
"<Nagdumán>"<br />
"dumán" vblex past @root #1->0<br />
"<akó>"<br />
"akó" prn p1 sg @nsubj #2->1<br />
"<sa>"<br />
"sa" det def @det #3->4<br />
"<haróng>"<br />
"haróng" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<dakúla>"<br />
"dakúla" adj sg @amod #6->4<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
* Translating to "I left the big house", this sentence demonstrates the most common structure of adjectival modifiers in Bikol, where the noun is followed by a linking word and the adjective. Dakúla (big), has an amod relation to the noun haróng (house).<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Although syntactically this sentence is very similar to the previous example, with a noun followed by a linking word and an adjective, this demonstrates a different kind of adjectival modifier. The adjective "nakadukót" is derived from the verb stem "dukót", meaning to glue or attach. The "naka" prefix changes the word into an adjective, meaning attached or stuck on. The sentence translates to, "Remove the poster attached to the bridge."<br />
<br />
<br />
<br />
[[Category: sp17_UD]]<br />
[[Category: Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=Bikol/Universal_Dependencies&diff=4908Bikol/Universal Dependencies2017-04-26T19:02:32Z<p>Eortici1: </p>
<hr />
<div>This page documents Bikol Universal Dependencies. <br />
<br />
== Evaluation ==<br />
<br />
== Dependency relations ==<br />
<br />
=== obj ===<br />
The object relation relates a verb and its object. Bikol is a VSO language, so the object of the verb is usually found after the subject, like in the following two examples.<br />
<br />
<pre><br />
"<Iinulí>"<br />
"ulí" vblex pres @root #1->0<br />
"<ko>"<br />
"ko" prn p1 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<libró>"<br />
"libró" n @obj #4->1<br />
"<.>"<br />
"." sent @punct #5->1<br />
</pre><br />
* In this sentence, which translates to "I return a book", libró is the object and depends on the verb, iinulí.<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Similarly, pósters is the object of this sentence and depends on the verb, ipaháli.<br />
<br />
=== advmod ===<br />
An adverbial modifier of a word is an adverb or adverbial phrase that modifies a predicate or modifier word.<br />
<br />
<pre><br />
"<Patosón>"<br />
"patós" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pansít>"<br />
"pansít" n @obj #4->1<br />
"<,>"<br />
"," cm @punct #5->6<br />
"<tápos>"<br />
"tápos" cnjadv @conj #6->1<br />
"<,>"<br />
"," cm @punct #7->8<br />
"<darahón>"<br />
"dará" vblex inf @conj #8->1<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #9->8<br />
"<digdí>"<br />
"digdí" adv @advmod #10->8<br />
"<.>"<br />
"." sent @punct #11->1<br />
</pre><br />
* In this sentence, digdí (or here) is the adverbial modifier that modifies the verb darahón (the command form of 'bring'). <br />
<br />
<pre><br />
"<Tápos>"<br />
"tápos" adj sg @root #1->0<br />
"<na>"<br />
"na" adv @advmod #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<prográma>"<br />
"prográma" n @nsubj #4->1<br />
"<sa>"<br />
"sa" pr @case #5->6<br />
"<gym>"<br />
"gym" n @nmod #6->4<br />
"<?>"<br />
"?" sent @punct #7->1<br />
</pre><br />
*This sentence translates to "Is the program in the gym already done?" The predicate is an implied "is", so the root the adjective, tápos (done). The adverbial modifier na (already) describes this adjective.<br />
<br />
=== advcl ===<br />
According to the Universal Dependencies documentation, an adverbial clause modifier is a clause which modifies a verb or other predicate as a modifier and not as a core complement. In my corpora, this relation came up in if statements.<br />
<br />
<pre><br />
"<Kon>"<br />
"kon" cnjsub @advmod #1->2<br />
"<makuá>"<br />
"kuá" vblex inf @advcl #2->5<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #3->2<br />
"<,>"<br />
"," cm @punct #4->5<br />
"<itaó>"<br />
"taó" vblex inf @root #5->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #6->5<br />
"<saná>"<br />
"saná" adv @advmod #7->5<br />
"<sakó>"<br />
"sakó" prn p1 sg @iobj #8->5<br />
"<.>"<br />
"." sent @punct #9->5<br />
</pre><br />
* This sentence translates it to "If you find it, just give it to me." The first clause, "If you find it", is an adverbial clause modifier which modifies "just give it to me". Thus, the root of the adverbial clause modifier, makuá, has an advcl dependency on the root of the independent clause that it modifies, itaó.<br />
<br />
<pre><br />
"<Kon>"<br />
"Kon" cnjsub @advmod #1->2<br />
"<maglúto>"<br />
"lúto" vblex inf @advcl #2->7<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #3->2<br />
"<nin>"<br />
"nin" link @clf #4->5<br />
"<jéllo>"<br />
"jéllo" n @obj #5->2<br />
"<,>"<br />
"," cm @punct #6->7<br />
"<magkuá>"<br />
"kuá" vblex inf @root #7->0<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #8->7<br />
"<nin>"<br />
"nin" link @clf #9->10<br />
"<pláto>"<br />
"pláto" n @obj #10->7<br />
"<sagkód>"<br />
"sagkód" adj sg @cc #11->12 ! and<br />
"<kutsára>"<br />
"kutsára" n @conj #12->10<br />
"<.>"<br />
"." sent @punct #13->2<br />
</pre><br />
* This sentence essentially has the same structure as the previous example. The root of the adverbial clause modifier, maglúto, has an advcl dependency on magkuá, the root of the independent clause.<br />
<br />
=== obl ===<br />
The obl (oblique) relation is used for a nominal that serves as a non-core argument or adjunct. In Bikol, this typically means that it functions as an adverbial attached to a predicate.<br />
<br />
<pre><br />
"<Hanápon>"<br />
"hánap" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<siyá>"<br />
"siyá" prn p3 sg @obj #3->1<br />
"<sa>"<br />
"sa" pr @case #4->5<br />
"<cafeteria>"<br />
"cafeteria" n @obl #5->1<br />
"<.>"<br />
"." sent @punct #6->1<br />
</pre><br />
* In this sentence, meaning "Look for him/her <i>in the cafeteria</i>", cafeteria is a nominal serving as a non-core argument and modifying the verb, hanápon (to look). Therefore cafeteria has an obl relation to hanápon.<br />
<br />
<pre><br />
"<Sublión>"<br />
"sublí" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<si>"<br />
"si" det def @det #3->4<br />
"<bóla>"<br />
"bóla" n @obj #4->1<br />
"<ki>"<br />
"ki" pr @case #5->6<br />
"<Jacób>"<br />
"Jacób" cog @obl #6->1<br />
"<.>"<br />
"." sent @punct #7->1<br />
</pre><br />
*This sentence behaves similarly as the previous example, except now the nominal in the oblique relation is a proper name (Jacób). It means "Borrow the ball from Jacob".<br />
<br />
[[Category: sp17_UD]]<br />
[[Category: Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=Bikol/Universal_Dependencies&diff=4907Bikol/Universal Dependencies2017-04-26T18:47:53Z<p>Eortici1: /* advcl */</p>
<hr />
<div>This page documents Bikol Universal Dependencies. <br />
<br />
== Evaluation ==<br />
<br />
== Dependency relations ==<br />
<br />
=== obj ===<br />
The object relation relates a verb and its object. Bikol is a VSO language, so the object of the verb is usually found after the subject, like in the following two examples.<br />
<br />
<pre><br />
"<Iinulí>"<br />
"ulí" vblex pres @root #1->0<br />
"<ko>"<br />
"ko" prn p1 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<libró>"<br />
"libró" n @obj #4->1<br />
"<.>"<br />
"." sent @punct #5->1<br />
</pre><br />
* In this sentence, which translates to "I return a book", libró is the object and depends on the verb, iinulí.<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Similarly, pósters is the object of this sentence and depends on the verb, ipaháli.<br />
<br />
=== advmod ===<br />
An adverbial modifier of a word is an adverb or adverbial phrase that modifies a predicate or modifier word.<br />
<br />
<pre><br />
"<Patosón>"<br />
"patós" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pansít>"<br />
"pansít" n @obj #4->1<br />
"<,>"<br />
"," cm @punct #5->6<br />
"<tápos>"<br />
"tápos" cnjadv @conj #6->1<br />
"<,>"<br />
"," cm @punct #7->8<br />
"<darahón>"<br />
"dará" vblex inf @conj #8->1<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #9->8<br />
"<digdí>"<br />
"digdí" adv @advmod #10->8<br />
"<.>"<br />
"." sent @punct #11->1<br />
</pre><br />
* In this sentence, digdí (or here) is the adverbial modifier that modifies the verb darahón (the command form of 'bring'). <br />
<br />
<pre><br />
"<Tápos>"<br />
"tápos" adj sg @root #1->0<br />
"<na>"<br />
"na" adv @advmod #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<prográma>"<br />
"prográma" n @nsubj #4->1<br />
"<sa>"<br />
"sa" pr @case #5->6<br />
"<gym>"<br />
"gym" n @nmod #6->4<br />
"<?>"<br />
"?" sent @punct #7->1<br />
</pre><br />
*This sentence translates to "Is the program in the gym already done?" The predicate is an implied "is", so the root the adjective, tápos (done). The adverbial modifier na (already) describes this adjective.<br />
<br />
=== advcl ===<br />
According to the Universal Dependencies documentation, an adverbial clause modifier is a clause which modifies a verb or other predicate as a modifier and not as a core complement. In my corpora, this relation came up in if statements.<br />
<br />
<pre><br />
"<Kon>"<br />
"kon" cnjsub @advmod #1->2<br />
"<makuá>"<br />
"kuá" vblex inf @advcl #2->5<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #3->2<br />
"<,>"<br />
"," cm @punct #4->5<br />
"<itaó>"<br />
"taó" vblex inf @root #5->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #6->5<br />
"<saná>"<br />
"saná" adv @advmod #7->5<br />
"<sakó>"<br />
"sakó" prn p1 sg @iobj #8->5<br />
"<.>"<br />
"." sent @punct #9->5<br />
</pre><br />
* This sentence translates it to "If you find it, just give it to me." The first clause, "If you find it", is an adverbial clause modifier which modifies "just give it to me". Thus, the root of the adverbial clause modifier, makuá, has an advcl dependency on the root of the independent clause that it modifies, itaó.<br />
<br />
<pre><br />
"<Kon>"<br />
"Kon" cnjsub @advmod #1->2<br />
"<maglúto>"<br />
"lúto" vblex inf @advcl #2->7<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #3->2<br />
"<nin>"<br />
"nin" link @clf #4->5<br />
"<jéllo>"<br />
"jéllo" n @obj #5->2<br />
"<,>"<br />
"," cm @punct #6->7<br />
"<magkuá>"<br />
"kuá" vblex inf @root #7->0<br />
"<kan>"<br />
"kan" prn p2 sg @nsubj #8->7<br />
"<nin>"<br />
"nin" link @clf #9->10<br />
"<pláto>"<br />
"pláto" n @obj #10->7<br />
"<sagkód>"<br />
"sagkód" adj sg @cc #11->12 ! and<br />
"<kutsára>"<br />
"kutsára" n @conj #12->10<br />
"<.>"<br />
"." sent @punct #13->2<br />
</pre><br />
* This sentence essentially has the same structure as the previous example. The root of the adverbial clause modifier, maglúto, has an advcl dependency on magkuá, the root of the independent clause.<br />
<br />
[[Category: sp17_UD]]<br />
[[Category: Bikol]]</div>Eortici1http://wikis.swarthmore.edu/ling073/index.php?title=Bikol/Universal_Dependencies&diff=4905Bikol/Universal Dependencies2017-04-26T18:21:00Z<p>Eortici1: /* advcl */</p>
<hr />
<div>This page documents Bikol Universal Dependencies. <br />
<br />
== Evaluation ==<br />
<br />
== Dependency relations ==<br />
<br />
=== obj ===<br />
The object relation relates a verb and its object. Bikol is a VSO language, so the object of the verb is usually found after the subject, like in the following two examples.<br />
<br />
<pre><br />
"<Iinulí>"<br />
"ulí" vblex pres @root #1->0<br />
"<ko>"<br />
"ko" prn p1 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<libró>"<br />
"libró" n @obj #4->1<br />
"<.>"<br />
"." sent @punct #5->1<br />
</pre><br />
* In this sentence, which translates to "I return a book", libró is the object and depends on the verb, iinulí.<br />
<br />
<pre><br />
"<Ipaháli>"<br />
"háli" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pósters>"<br />
"pósters" n @obj #4->1<br />
"<na>"<br />
"na" link @clf #5->6<br />
"<nakadukót>"<br />
"dukót" adj sg @amod #6->4<br />
"<sa>"<br />
"sa" pr @case #7->8<br />
"<tuláy>"<br />
"tuláy" n @obl #8->1<br />
"<..>"<br />
".." sent @punct #9->1<br />
</pre><br />
* Similarly, pósters is the object of this sentence and depends on the verb, ipaháli.<br />
<br />
=== advmod ===<br />
An adverbial modifier of a word is an adverb or adverbial phrase that modifies a predicate or modifier word.<br />
<br />
<pre><br />
"<Patosón>"<br />
"patós" vblex inf @root #1->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<pansít>"<br />
"pansít" n @obj #4->1<br />
"<,>"<br />
"," cm @punct #5->6<br />
"<tápos>"<br />
"tápos" cnjadv @conj #6->1<br />
"<,>"<br />
"," cm @punct #7->8<br />
"<darahón>"<br />
"dará" vblex inf @conj #8->1<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #9->8<br />
"<digdí>"<br />
"digdí" adv @advmod #10->8<br />
"<.>"<br />
"." sent @punct #11->1<br />
</pre><br />
* In this sentence, digdí (or here) is the adverbial modifier that modifies the verb darahón (the command form of 'bring'). <br />
<br />
<pre><br />
"<Tápos>"<br />
"tápos" adj sg @root #1->0<br />
"<na>"<br />
"na" adv @advmod #2->1<br />
"<an>"<br />
"an" det ind @det #3->4<br />
"<prográma>"<br />
"prográma" n @nsubj #4->1<br />
"<sa>"<br />
"sa" pr @case #5->6<br />
"<gym>"<br />
"gym" n @nmod #6->4<br />
"<?>"<br />
"?" sent @punct #7->1<br />
</pre><br />
*This sentence translates to "Is the program in the gym already done?" The predicate is an implied "is", so the root the adjective, tápos (done). The adverbial modifier na (already) describes this adjective.<br />
<br />
=== advcl ===<br />
According to the Universal Dependencies documentation, an adverbial clause modifier is a clause which modifies a verb or other predicate as a modifier and not as a core complement. In my corpora, this relation came up in if statements.<br />
<br />
<pre><br />
"<Kon>"<br />
"kon" cnjsub @advmod #1->2<br />
"<makuá>"<br />
"kuá" vblex inf @advcl #2->5<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #3->2<br />
"<,>"<br />
"," cm @punct #4->5<br />
"<itaó>"<br />
"taó" vblex inf @root #5->0<br />
"<mo>"<br />
"mo" prn p2 sg @nsubj #6->5<br />
"<saná>"<br />
"saná" adv @advmod #7->5<br />
"<sakó>"<br />
"sakó" prn p1 sg @iobj #8->5<br />
"<.>"<br />
"." sent @punct #9->5<br />
</pre><br />
* Comment<br />
<br />
[[Category: sp17_UD]]<br />
[[Category: Bikol]]</div>Eortici1