http://wikis.swarthmore.edu/ling073/api.php?action=feedcontributions&feedformat=atom&user=Jmalin1LING073 - User contributions [en]2026-11-14T03:44:05ZUser contributionsMediaWiki 1.27.7http://wikis.swarthmore.edu/ling073/index.php?title=User:Jmalin1/Final_Project&diff=5395User:Jmalin1/Final Project2017-05-11T02:36:07Z<p>Jmalin1: </p>
<hr />
<div>=Transducer for Innu=<br />
For my final project I began a transducer for Innu (also known as Montagnais). It is available on [https://github.swarthmore.edu/jmalin1/ling073-moe Swarthmore's github] as well as [https://github.com/jsrmalin/ling073-moe git].<br />
<br />
==Implementation==<br />
The transducer handles the following major features of Innu morphology<br />
<br />
===Noun Possession===<br />
Possessed nouns in Innu receive a circumfix indicating the number and person of the possessor, and sometimes also a suffix -(i)m indicating possession. The transducer currently applies both pieces of morphology, but the appearance of the possession suffix is influenced by a number of factors including phonology, the gender of the noun, and whether or not the possession is alienable. For example, "my head" would be nustikuan in reference to one's own head and nustikuanim in reference to the detached head of an animal that the speaker owns. Ensuring that this suffix is conditioned correctly in the transducer will require further study of Innu texts or consultation with a speaker.<br />
<br />
===Verb Subject and Object Agreement===<br />
Innu transitive verbs are inflected for both subject and object. The "subject" of the verb is marked as whichever participant is higher in the hierarchy of person: 2>1>3>3(obv). If the subject of the action is lower in the hierarchy, then the verb is placed in an "inverted" form with a different suffix and sometimes a stem change. For example, using the verb petu (hear):<br />
*tshipetun ("You hear me") is marked with the second person prefix and first person suffix.<br />
*tshipetatin ("I hear you") is still marked with the second person prefix, but a different first person suffix indicating the the first person is the subject.<br />
<br />
==Evaluation==<br />
Coverage was tested over a preliminary corpus (8476 characters without spaces) composed of 4 myths taken from the [http://www.innu-aimun.ca/english/stories/ Innu-Aimun website]. The corpus is available in a repository on [https://github.swarthmore.edu/jmalin1/ling073-moe-corpus Swarthmore's github], which is private due to copyright.<br />
<br />
Current corpus coverage is 54.58% (although approximately 22% of analyzed tokens are punctuation).<br />
<br />
==Next Steps==<br />
<br />
===Double Obviation===<br />
Suffixes for obviation may change when an obviative noun is possessed by another obviative noun. Currently, this is implemented in the transducer, but messily, with some overgeneration of forms for possessed obviative nouns. It should be updated with greater understanding of the morphological processes involved.<br />
<br />
===More Verb Paradigms===<br />
While all major verb paradigms have been implemented in the present indicative, other tenses and moods have yet to be implemented. The future tense in particular has the issue that morphemes generally analyzed as verb prefixes are instead written as a separate word, meaning some reformatting of the other verb paradigms may be required to accommodate this.<br />
<br />
===Disambiguation===<br />
There are several ambiguous stems in the transducer - for example, many Innu adverbs are also interjections or conjunctions with different meanings. Variable word order means that precise disambiguation rules are difficult to come up with.<br />
<br />
===Noun and Verb Formation===<br />
As a polysynthetic language, Innu has several productive derivational processes. Currently, these are handled with different stems in the transducer, but future versions of the transducer could perhaps handle them automatically.<br />
<br />
===Dialectical Variation===<br />
Innu has several different dialects. The transducer functions in a modern, standardized orthography but could be modified to analyze text written in variant orthographies as well.<br />
<br />
[[Category:sp17_FinalProjects]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=User:Jmalin1/Final_Project&diff=5391User:Jmalin1/Final Project2017-05-11T02:07:01Z<p>Jmalin1: /* Verb Subject and Object Agreement */</p>
<hr />
<div>=Transducer for Innu=<br />
For my final project I began a transducer for Innu (also known as Montagnais). It is available on [https://github.swarthmore.edu/jmalin1/ling073-moe Swarthmore's github].<br />
<br />
==Implementation==<br />
The transducer handles the following major features of Innu morphology<br />
<br />
===Noun Possession===<br />
Possessed nouns in Innu receive a circumfix indicating the number and person of the possessor, and sometimes also a suffix -(i)m indicating possession. The transducer currently applies both pieces of morphology, but the appearance of the possession suffix is influenced by a number of factors including phonology, the gender of the noun, and whether or not the possession is alienable. For example, "my head" would be nustikuan in reference to one's own head and nustikuanim in reference to the detached head of an animal that the speaker owns. Ensuring that this suffix is conditioned correctly in the transducer will require further study of Innu texts or consultation with a speaker.<br />
<br />
===Verb Subject and Object Agreement===<br />
Innu transitive verbs are inflected for both subject and object. The "subject" of the verb is marked as whichever participant is higher in the hierarchy of person: 2>1>3>3(obv). If the subject of the action is lower in the hierarchy, then the verb is placed in an "inverted" form with a different suffix and sometimes a stem change. For example, using the verb petu (hear):<br />
*tshipetun ("You hear me") is marked with the second person prefix and first person suffix.<br />
*tshipetatin ("I hear you") is still marked with the second person prefix, but a different first person suffix indicating the the first person is the subject.<br />
<br />
==Evaluation==<br />
Coverage was tested over a preliminary corpus (8476 characters without spaces) composed of 4 myths taken from the [http://www.innu-aimun.ca/english/stories/ Innu-Aimun website]. The corpus is available in a repository on [https://github.swarthmore.edu/jmalin1/ling073-moe-corpus Swarthmore's github], which is private due to copyright.<br />
<br />
Current corpus coverage is 54.58% (although approximately 22% of analyzed tokens are punctuation).<br />
<br />
==Next Steps==<br />
<br />
===Double Obviation===<br />
Suffixes for obviation may change when an obviative noun is possessed by another obviative noun. Currently, this is implemented in the transducer, but messily, with some overgeneration of forms for possessed obviative nouns. It should be updated with greater understanding of the morphological processes involved.<br />
<br />
===More Verb Paradigms===<br />
While all major verb paradigms have been implemented in the present indicative, other tenses and moods have yet to be implemented. The future tense in particular has the issue that morphemes generally analyzed as verb prefixes are instead written as a separate word, meaning some reformatting of the other verb paradigms may be required to accommodate this.<br />
<br />
===Disambiguation===<br />
There are several ambiguous stems in the transducer - for example, many Innu adverbs are also interjections or conjunctions with different meanings. Variable word order means that precise disambiguation rules are difficult to come up with.<br />
<br />
===Noun and Verb Formation===<br />
As a polysynthetic language, Innu has several productive derivational processes. Currently, these are handled with different stems in the transducer, but future versions of the transducer could perhaps handle them automatically.<br />
<br />
===Dialectical Variation===<br />
Innu has several different dialects. The transducer functions in a modern, standardized orthography but could be modified to analyze text written in variant orthographies as well.<br />
<br />
[[Category:sp17_FinalProjects]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=User:Jmalin1/Final_Project&diff=5390User:Jmalin1/Final Project2017-05-11T02:06:44Z<p>Jmalin1: </p>
<hr />
<div>=Transducer for Innu=<br />
For my final project I began a transducer for Innu (also known as Montagnais). It is available on [https://github.swarthmore.edu/jmalin1/ling073-moe Swarthmore's github].<br />
<br />
==Implementation==<br />
The transducer handles the following major features of Innu morphology<br />
<br />
===Noun Possession===<br />
Possessed nouns in Innu receive a circumfix indicating the number and person of the possessor, and sometimes also a suffix -(i)m indicating possession. The transducer currently applies both pieces of morphology, but the appearance of the possession suffix is influenced by a number of factors including phonology, the gender of the noun, and whether or not the possession is alienable. For example, "my head" would be nustikuan in reference to one's own head and nustikuanim in reference to the detached head of an animal that the speaker owns. Ensuring that this suffix is conditioned correctly in the transducer will require further study of Innu texts or consultation with a speaker.<br />
<br />
===Verb Subject and Object Agreement===<br />
Innu transitive verbs are inflected for both subject and object. The "subject" of the verb is marked as whichever participant is higher in the hierarchy of person: 2>1>3>3(obv). If the subject of the action is lower in the hierarchy, then the verb is placed in an "inverted" form with a different suffix and sometimes a stem change. For example, using the verb petu (hear):<br />
**tshipetun ("You hear me") is marked with the second person prefix and first person suffix.<br />
**tshipetatin ("I hear you") is still marked with the second person prefix, but a different first person suffix indicating the the first person is the subject.<br />
<br />
==Evaluation==<br />
Coverage was tested over a preliminary corpus (8476 characters without spaces) composed of 4 myths taken from the [http://www.innu-aimun.ca/english/stories/ Innu-Aimun website]. The corpus is available in a repository on [https://github.swarthmore.edu/jmalin1/ling073-moe-corpus Swarthmore's github], which is private due to copyright.<br />
<br />
Current corpus coverage is 54.58% (although approximately 22% of analyzed tokens are punctuation).<br />
<br />
==Next Steps==<br />
<br />
===Double Obviation===<br />
Suffixes for obviation may change when an obviative noun is possessed by another obviative noun. Currently, this is implemented in the transducer, but messily, with some overgeneration of forms for possessed obviative nouns. It should be updated with greater understanding of the morphological processes involved.<br />
<br />
===More Verb Paradigms===<br />
While all major verb paradigms have been implemented in the present indicative, other tenses and moods have yet to be implemented. The future tense in particular has the issue that morphemes generally analyzed as verb prefixes are instead written as a separate word, meaning some reformatting of the other verb paradigms may be required to accommodate this.<br />
<br />
===Disambiguation===<br />
There are several ambiguous stems in the transducer - for example, many Innu adverbs are also interjections or conjunctions with different meanings. Variable word order means that precise disambiguation rules are difficult to come up with.<br />
<br />
===Noun and Verb Formation===<br />
As a polysynthetic language, Innu has several productive derivational processes. Currently, these are handled with different stems in the transducer, but future versions of the transducer could perhaps handle them automatically.<br />
<br />
===Dialectical Variation===<br />
Innu has several different dialects. The transducer functions in a modern, standardized orthography but could be modified to analyze text written in variant orthographies as well.<br />
<br />
[[Category:sp17_FinalProjects]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=User:Jmalin1/Final_Project&diff=5386User:Jmalin1/Final Project2017-05-11T01:48:25Z<p>Jmalin1: </p>
<hr />
<div>=Transducer for Innu=<br />
For my final project I began a transducer for Innu (also known as Montagnais). It is available on [https://github.swarthmore.edu/jmalin1/ling073-moe Swarthmore's github].<br />
<br />
==Evaluation==<br />
Coverage was tested over a preliminary corpus (8476 characters without spaces) composed of 4 myths taken from the [http://www.innu-aimun.ca/english/stories/ Innu-Aimun website]. The corpus is available in a repository on [https://github.swarthmore.edu/jmalin1/ling073-moe-corpus Swarthmore's github], which is private due to copyright.<br />
<br />
Current corpus coverage is 54.58% (although approximately 22% of analyzed tokens are punctuation).<br />
<br />
==Next Steps==<br />
<br />
===Double Obviation===<br />
Suffixes for obviation may change when an obviative noun is possessed by another obviative noun. Currently, this is implemented in the transducer, but messily, with some overgeneration of forms for possessed obviative nouns. It should be updated with greater understanding of the morphological processes involved.<br />
<br />
===More Verb Paradigms===<br />
While all major verb paradigms have been implemented in the present indicative, other tenses and moods have yet to be implemented. The future tense in particular has the issue that morphemes generally analyzed as verb prefixes are instead written as a separate word, meaning some reformatting of the other verb paradigms may be required to accommodate this.<br />
<br />
===Disambiguation===<br />
There are several ambiguous stems in the transducer - for example, many Innu adverbs are also interjections or conjunctions with different meanings. Variable word order means that precise disambiguation rules are difficult to come up with.<br />
<br />
===Noun and Verb Formation===<br />
As a polysynthetic language, Innu has several productive derivational processes. Currently, these are handled with different stems in the transducer, but future versions of the transducer could perhaps handle them automatically.<br />
<br />
===Dialectical Variation===<br />
Innu has several different dialects. The transducer functions in a modern, standardized orthography but could be modified to analyze text written in variant orthographies as well.<br />
<br />
[[Category:sp17_FinalProjects]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=User:Jmalin1/Final_Project&diff=5385User:Jmalin1/Final Project2017-05-11T01:40:25Z<p>Jmalin1: </p>
<hr />
<div>=Transducer for Innu=<br />
For my final project I began a transducer for Innu (also known as Montagnais). It is available on [https://github.swarthmore.edu/jmalin1/ling073-moe Swarthmore's github].<br />
<br />
==Evaluation==<br />
Coverage was tested over a preliminary corpus (8476 characters without spaces) composed of 4 myths taken from the [http://www.innu-aimun.ca/english/stories/ Innu-Aimun website]. The corpus is available in a repository on [https://github.swarthmore.edu/jmalin1/ling073-moe-corpus Swarthmore's github], which is private due to copyright.<br />
<br />
Current corpus coverage is 54.58% (although approximately 22% of analyzed tokens are punctuation).<br />
<br />
==Next Steps==<br />
<br />
===Double Obviation===<br />
Suffixes for obviation may change when an obviative noun is possessed by another obviative noun. Currently, this is implemented in the transducer, but messily, with some overgeneration of forms for possessed obviative nouns. It should be updated with greater understanding of the morphological processes involved.<br />
<br />
===More Verb Paradigms===<br />
While all major verb paradigms have been implemented in the present indicative, other tenses and moods have yet to be implemented. The future tense in particular has the issue that morphemes generally analyzed as verb prefixes are instead written as a separate word, meaning some reformatting of the other verb paradigms may be required to accommodate this.<br />
<br />
===Disambiguation===<br />
There are several ambiguous stems in the transducer - for example, many Innu adverbs are also interjections or conjunctions with different meanings. Variable word order means that precise disambiguation rules are difficult to come up with.<br />
<br />
===Noun and Verb Formation===<br />
As a polysynthetic language, Innu has several productive derivational processes. Currently, these are handled with different stems in the transducer, but future versions of the transducer could perhaps handle them automatically.<br />
<br />
===Dialectical Variation===<br />
Innu has several different dialects. The transducer functions in a modern, standardized orthography but could be modified to analyze text written in variant orthographies as well.</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=User:Jmalin1/Final_Project&diff=5384User:Jmalin1/Final Project2017-05-11T01:38:02Z<p>Jmalin1: /* Next Steps */</p>
<hr />
<div>=Transducer for Innu=<br />
For my final project I began a transducer for Innu (also known as Montagnais). It is available on [https://github.swarthmore.edu/jmalin1/ling073-moe Swarthmore's github].<br />
<br />
==Evaluation==<br />
Coverage was tested over a preliminary corpus (8476 characters without spaces) composed of 4 myths taken from the [http://www.innu-aimun.ca/english/stories/ Innu-Aimun website]. The corpus is available in a repository on [https://github.swarthmore.edu/jmalin1/ling073-moe-corpus Swarthmore's github], which is private due to copyright.<br />
<br />
Current corpus coverage is 54.58% (although approximately 22% of analyzed tokens are punctuation).<br />
<br />
==Next Steps==<br />
<br />
===Double Obviation===<br />
Suffixes for obviation may change when an obviative noun is possessed by another obviative noun. Currently, this is implemented in the transducer, but messily, with some overgeneration of forms for possessed obviative nouns. It should be updated with greater understanding of the morphological processes involved.<br />
<br />
===More Verb Paradigms===<br />
While all major verb paradigms have been implemented in the present indicative, other tenses and moods have yet to be implemented. The future tense in particular has the issue that morphemes generally analyzed as verb prefixes are instead written as a separate word, meaning some reformatting of the other verb paradigms may be required to accommodate this.<br />
<br />
===Disambiguation===<br />
There are several ambiguous stems in the transducer - for example, many Innu adverbs are also interjections or conjunctions with different meanings. Variable word order means that precise disambiguation rules are difficult to come up with.<br />
<br />
===Noun and Verb Formation===<br />
As a polysynthetic language, Innu has several productive derivational processes. Currently, these are handled with different stems in the transducer, but future versions of the transducer could perhaps handle them automatically.<br />
<br />
==Dialectical Variation==<br />
Innu has several different dialects. The transducer functions in a modern, standardized orthography but could be modified to analyze text written in variant orthographies as well.</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=User:Jmalin1/Final_Project&diff=5383User:Jmalin1/Final Project2017-05-11T01:33:51Z<p>Jmalin1: Created page with "=Transducer for Innu= For my final project I began a transducer for Innu (also known as Montagnais). It is available on [https://github.swarthmore.edu/jmalin1/ling073-moe Swa..."</p>
<hr />
<div>=Transducer for Innu=<br />
For my final project I began a transducer for Innu (also known as Montagnais). It is available on [https://github.swarthmore.edu/jmalin1/ling073-moe Swarthmore's github].<br />
<br />
==Evaluation==<br />
Coverage was tested over a preliminary corpus (8476 characters without spaces) composed of 4 myths taken from the [http://www.innu-aimun.ca/english/stories/ Innu-Aimun website]. The corpus is available in a repository on [https://github.swarthmore.edu/jmalin1/ling073-moe-corpus Swarthmore's github], which is private due to copyright.<br />
<br />
Current corpus coverage is 54.58% (although approximately 22% of analyzed tokens are punctuation).<br />
<br />
==Next Steps==<br />
<br />
===Double Obviation===<br />
Suffixes for obviation may change when an obviative noun is possessed by another obviative noun. Currently, this is implemented in the transducer, but messily, with some overgeneration of forms for possessed obviative nouns. It should be updated with greater understanding of the morphological processes involved.<br />
<br />
===More Verb Paradigms===<br />
While all major verb paradigms have been implemented in the present indicative, other tenses and moods have yet to be implemented. The future tense in particular has the issue that morphemes generally analyzed as verb prefixes are instead written as a separate word, meaning some reformatting of the other verb paradigms may be required to accommodate this.<br />
<br />
===Disambiguation===<br />
There are several ambiguous stems in the transducer - for example, many Innu adverbs are also interjections or conjunctions with different meanings. Variable word order means that disambiguation rules are difficult to write.<br />
<br />
===Noun and Verb Formation===<br />
As a polysynthetic language, Innu has several productive derivational processes. Currently, these are handled with different stems in the transducer, but future versions of the transducer could perhaps handle them automatically.</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak/Universal_Dependencies&diff=4867Wôpanâak/Universal Dependencies2017-04-25T03:28:55Z<p>Jmalin1: /* Dependency Relations */</p>
<hr />
<div>==Evaluation==<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation<br />
|-<br />
! <br />
! wam.withmorph.udpipe<br />
! wam.nomorph.udpipe<br />
! Number of forms<br />
! Number of sentences<br />
|-<br />
! wam.annotated.ud.conllu<br />
| UAS: 98.57%, LAS: 96.43%<br />
| UAS: 88.57%, LAS: 88.57%<br />
| 140<br />
| 95<br />
|-<br />
! wam.annotated2.ud.conllu<br />
| UAS: 43.75%, LAS: 23.96%<br />
| UAS: 38.54%, LAS: 23.96%<br />
| 96<br />
| 25<br />
|}<br />
<br />
The lower scores for the second corpus are expected because the first corpus, on which the parser was trained, is composed of extremely simple sentences from a grammar while the second corpus contains some more complex sentences from the Bible in addition to simple sentences.<br />
<br />
==Dependency Relations==<br />
<br />
===nsubj===<br />
The nsubj relation is used to relate a noun that is the subject of a clause to the verb that heads the clause. In Wôpanâak, the nsubj relation occurs in any sentence with a subject expressed separately from a verb. This does not occur in every sentence, as sometimes subject information is carried entirely within the verb.<br />
Examples include:<br />
* "masq qaqeew": In this simple sentence meaning "The bear is running", "masq" has an nsubj relation with "qaqeew".<br />
* "ôk nȣwâw God weeqayach": In this clause meaning "So God says 'let there be light'", "God" has an nsubj relation with "nȣwâw".<br />
<br />
===obj===<br />
The obj relation relates a noun that is the object of a verb to the verb. In Wôpanâak, this relation is commonly used in sentences with transitive verbs, as long as the object is expressed as a separate word and not as a marking on the verb.<br />
Examples include:<br />
* "anum nâwâw masqah": This sentence means "The dog sees the bear". In it, "masq" has an obj relation with nâwâw.<br />
* "ayum God keesuq kah ahkee": In this clause meaning "God creates the sky and the earth", "keesuq" has an obj relation with "ayum".<br />
<br />
===obl===<br />
The obl relation is used to relate a noun that is a non-subject, non-object constituent of its clause to the head of the clause. In Wôpanâak, this relation is often used in sentences that express location or have nouns otherwise in the locative form.<br />
Examples include:<br />
* "nutaquneônus nȣshut": This sentence means "I am like my father". Since "aquneônus" is an intransitive verb, "nȣshut" is in the locative and has an obl relation with it.<br />
* "pôkeenum wôskeechee mȣnô": In this clause meaning "darkness was over the deep", "mȣnô" has an obl relation with "wôskeechee", the adverb acting as the head of the clause since the clause is copular.<br />
<br />
===conj===<br />
The conj relation is used to link an element coordinated to an earlier element to that element. In Wôpanâak it is used for nouns and clauses linked by the coordinating conjunction "kah".<br />
Examples include:<br />
* "ayum God keesuq kah ahkee": Again in "God created the sky and the earth", "ahkee" is linked to "keesuq" through a conj relation.<br />
* "pôkeenum wôskeechee mȣnô kah nashâanut wôskeechee nupeekôtȣ": In this clause, "darkness was over the deep and the spirit of God was over the water", the second instance of "wôskeechee" is linked to the first through a conj relation. This is because both are the heads of their respective copular clauses, which are linked by "kah".<br />
<br />
===discourse===<br />
The discourse relation is used to relate a word with a function relating to the discourse of a sentence (but otherwise lacking a role in the sentence) to the head of the sentence. In Wôpanâak it can be used to relate clause-beginning words like "kah" ("and") or "ôk" ("so") to the head of the clause.<br />
Examples include:<br />
* "nukees, kutahtawun.": This sentence means "Yes, you have it." The word "nukees" ("yes") is linked to the verb "kuhtahtawun" through the discourse relation.<br />
* "kah wutusôweetamun God weeqay keesuqôt": This means "And God called the light day". "Kah" is linked to "wutusôweetamun" with the discourse relation since it serves no purpose in the sentence other than to introduce the sentence.</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak/Universal_Dependencies&diff=4866Wôpanâak/Universal Dependencies2017-04-25T03:22:17Z<p>Jmalin1: /* Dependency Relations */</p>
<hr />
<div>==Evaluation==<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation<br />
|-<br />
! <br />
! wam.withmorph.udpipe<br />
! wam.nomorph.udpipe<br />
! Number of forms<br />
! Number of sentences<br />
|-<br />
! wam.annotated.ud.conllu<br />
| UAS: 98.57%, LAS: 96.43%<br />
| UAS: 88.57%, LAS: 88.57%<br />
| 140<br />
| 95<br />
|-<br />
! wam.annotated2.ud.conllu<br />
| UAS: 43.75%, LAS: 23.96%<br />
| UAS: 38.54%, LAS: 23.96%<br />
| 96<br />
| 25<br />
|}<br />
<br />
The lower scores for the second corpus are expected because the first corpus, on which the parser was trained, is composed of extremely simple sentences from a grammar while the second corpus contains some more complex sentences from the Bible in addition to simple sentences.<br />
<br />
==Dependency Relations==<br />
<br />
===nsubj===<br />
The nsubj relation is used to relate a noun that is the subject of a clause to the verb that heads the clause. In Wôpanâak, the nsubj relation occurs in any sentence with a subject expressed separately from a verb. This does not occur in every sentence, as sometimes subject information is carried entirely within the verb.<br />
Examples include:<br />
* "masq qaqeew": In this simple sentence meaning "The bear is running", "masq" has an nsubj relation with "qaqeew".<br />
* "ôk nȣwâw God weeqayach": In this clause meaning "So God says 'let there be light'", "God" has an nsubj relation with "nȣwâw".<br />
<br />
===obj===<br />
The obj relation relates a noun that is the object of a verb to the verb. In Wôpanâak, this relation is commonly used in sentences with transitive verbs, as long as the object is expressed as a separate word and not as a marking on the verb.<br />
Examples include:<br />
* "anum nâwâw masqah": This sentence means "The dog sees the bear". In it, "masq" has an obj relation with nâwâw.<br />
* "ayum God keesuq kah ahkee": In this clause meaning "God creates the sky and the earth", "keesuq" has an obj relation with "ayum".<br />
<br />
===obl===<br />
The obl relation is used to relate a noun that is a non-subject, non-object constituent of its clause to the head of the clause. In Wôpanâak, this relation is often used in sentences that express location or have nouns otherwise in the locative form.<br />
Examples include:<br />
* "nutaquneônus nȣshut": This sentence means "I am like my father". Since "aquneônus" is an intransitive verb, "nȣshut" is in the locative and has an obl relation with it.<br />
* "pôkeenum wôskeechee mȣnô": In this clause meaning "darkness was over the deep", "mȣnô" has an obl relation with "wôskeechee", the adverb acting as the head of the clause since the clause is copular.<br />
<br />
===discourse===<br />
The discourse relation is used to relate a word with a function relating to the discourse of a sentence (but otherwise lacking a role in the sentence) to the head of the sentence. In Wôpanâak it can be used to relate clause-beginning words like "kah" ("and") or "ôk" ("so") to the head of the clause.<br />
Examples include:<br />
* "nukees, kutahtawun.": This sentence means "Yes, you have it." The word "nukees" ("yes") is linked to the verb "kuhtahtawun" through the discourse relation.<br />
* "kah wutusôweetamun God weeqay keesuqôt": This means "And God called the light day". "Kah" is linked to "wutusôweetamun" with the discourse relation since it serves no purpose in the sentence other than to introduce the sentence.</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak/Universal_Dependencies&diff=4863Wôpanâak/Universal Dependencies2017-04-25T02:52:26Z<p>Jmalin1: </p>
<hr />
<div>==Evaluation==<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation<br />
|-<br />
! <br />
! wam.withmorph.udpipe<br />
! wam.nomorph.udpipe<br />
! Number of forms<br />
! Number of sentences<br />
|-<br />
! wam.annotated.ud.conllu<br />
| UAS: 98.57%, LAS: 96.43%<br />
| UAS: 88.57%, LAS: 88.57%<br />
| 140<br />
| 95<br />
|-<br />
! wam.annotated2.ud.conllu<br />
| UAS: 43.75%, LAS: 23.96%<br />
| UAS: 38.54%, LAS: 23.96%<br />
| 96<br />
| 25<br />
|}<br />
<br />
The lower scores for the second corpus are expected because the first corpus, on which the parser was trained, is composed of extremely simple sentences from a grammar while the second corpus contains some more complex sentences from the Bible in addition to simple sentences.<br />
<br />
==Dependency Relations==</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak/Universal_Dependencies&diff=4862Wôpanâak/Universal Dependencies2017-04-25T02:51:54Z<p>Jmalin1: </p>
<hr />
<div>==Evaluation==<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation<br />
|-<br />
! <br />
! wam.withmorph.udpipe<br />
! wam.nomorph.udpipe<br />
! Number of forms<br />
! Number of sentences<br />
|-<br />
! wam.annotated.ud.conllu<br />
| UAS: 98.57%, LAS: 96.43%<br />
| UAS: 88.57%, LAS: 88.57%<br />
| 140<br />
| 95<br />
|-<br />
! wam.annotated2.ud.conllu<br />
| UAS: 43.75%, LAS: 23.96%<br />
| UAS: 38.54%, LAS: 23.96%<br />
| 96<br />
| 25<br />
|}<br />
<br />
The lower scores for the second corpus are expected because the first corpus, on which the parser was trained, is composed of extremely simple sentences from a grammar while the second corpus contains some more complex sentences from the Bible in addition to simple sentences.</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak/Universal_Dependencies&diff=4861Wôpanâak/Universal Dependencies2017-04-25T02:51:13Z<p>Jmalin1: </p>
<hr />
<div>==Evaluation==<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation<br />
|-<br />
! <br />
! wam.withmorph.udpipe<br />
! wam.nomorph.udpipe<br />
! Number of forms<br />
! Number of sentences<br />
|-<br />
! wam.annotated.ud.conllu<br />
| UAS: 98.57%, LAS: 96.43%<br />
| UAS: 88.57%, LAS: 88.57%<br />
| 140<br />
| 95<br />
|-<br />
! wam.annotated2.ud.conllu<br />
| UAS: 43.75%, LAS: 23.96%<br />
| UAS: 38.54%, LAS: 23.96%<br />
| 96<br />
| 25<br />
|}<br />
<br />
The lower scores for the second corpus is expected because the first corpus, on which the parser was trained, is composed of extremely simple sentences from a grammar while the second corpus contains some more complex sentences from the Bible.</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak/Universal_Dependencies&diff=4860Wôpanâak/Universal Dependencies2017-04-25T02:50:02Z<p>Jmalin1: Created page with "==Evaluation== {| class="wikitable" |+ UD Parser Evaluation |- ! ! wam.withmorph.udpipe ! wam.nomorph.udpipe ! Number of forms ! Number of sentences |- ! wam.annotated.ud.con..."</p>
<hr />
<div>==Evaluation==<br />
{| class="wikitable"<br />
|+ UD Parser Evaluation<br />
|-<br />
! <br />
! wam.withmorph.udpipe<br />
! wam.nomorph.udpipe<br />
! Number of forms<br />
! Number of sentences<br />
|-<br />
! wam.annotated.ud.conllu<br />
| UAS: 98.57%, LAS: 96.43%<br />
| UAS: 88.57%, LAS: 88.57%<br />
| 140<br />
| 95<br />
|-<br />
! wam.annotated2.ud.conllu<br />
| UAS: 43.75%, LAS: 23.96%<br />
| UAS: 38.54%, LAS: 23.96%<br />
| 96<br />
| 25<br />
|}</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak&diff=4859Wôpanâak2017-04-25T02:38:38Z<p>Jmalin1: /* Universal Dependency */</p>
<hr />
<div>==External Resources==<br />
<br />
===Grammar===<br />
<br />
A grammar of the language published in 2000 by Jessie Little Doe Fermino (or Jessie Little Doe Baird) as an MIT thesis is available online [http://dspace.mit.edu/handle/1721.1/8740 here]. Most other resources I have found at this point are the sources used to develop this grammar. It is under copyright.<br />
<br />
Another [http://dspace.mit.edu/handle/1721.1/37422 thesis], by Nitana Hicks, lists morphemes used in Wôpanâak verbs along with rules for verb construction.<br />
<br />
Additionally, Volume 2 of ''Native Writings in Massachusett'' contains a "Grammatical Sketch" based on several centuries-older grammars.<br />
<br />
===Dictionary===<br />
<br />
The Natick Dictionary, published in 1903, appears to be a dictionary of at least one form of the language available online [https://archive.org/details/natickdictionar02trumgoog here]. It is a dictionary between English and the Natick dialect of Wôpanâak.<br />
<br />
===Corpora===<br />
<br />
The 1988 book ''Native Writings In Massachusett'' (available from McCabe and online), by Ives Goddard and Kathleen Bragdon, is the primary source of text for the language, containing many (possibly all) surviving texts written by native speakers of the language from the 1660s to the 1750s. The book contains many documents with transcriptions, annotations, and translations. The documents are in large part legal documents, including land transactions and wills, although documents with other purposes are in the collection as well. The book was published by the American Philosophical Society and is copyrighted.<br />
<br />
Additionally, a major source for the language is a translation of the Bible written by a non-native speaker (missionary John Eliot). Editions from [https://archive.org/details/mamussewunneetup00elio 1663] and [https://archive.org/details/mamussewunneetup02elio 1685] are available online, although they appear to be in image format without good transcription.<br />
<br />
==Developed Resources==<br />
<br />
===Keyboard Layouts===<br />
<br />
Preliminary [[Wôpanâak/Keyboard|keyboard layouts]] for Wôpanâak are available on [https://github.swarthmore.edu/jmalin1/ling073-wam-keyboard GitHub].<br />
<br />
[[Category:Wôpanâak]] [[Category:Sp17 ResourceDocumentation]]<br />
<br />
===Corpora===<br />
<br />
Basic corpora are available [https://github.swarthmore.edu/jmalin1/ling073-wam-corpus/graphs/contributors here] (still being constructed). The repository is private due to copyrighted sources for the corpora.<br />
<br />
===Morphological Analyzer===<br />
<br />
Some morphological processes are documented [[Wôpanâak/Grammar|here]].<br />
<br />
A preliminary morphological analyzer is under development [https://github.swarthmore.edu/jmalin1/ling073-wam here]. Currently it mainly analyzes nouns. More information can be found at [[Wôpanâak/Transducer]].<br />
<br />
===Disambiguation===<br />
<br />
A disambiguator is under development at [[Wôpanâak/Disambiguation]].<br />
<br />
===Khasi and Wôpanâak===<br />
<br />
A page containing all resources devoted to language pair Khasi and Wôpanâak can be found here: [[Khasi_and_Wôpanâak|Khasi and Wôpanâak]]<br />
<br />
===Contrastive Grammars===<br />
<br />
Differences in grammar between Khasi and Wôpanâak can be found at [[Khasi_and_Wôpanâak/ContrastiveGrammars|Contrastive Grammars]]<br />
<br />
===Universal Dependency===<br />
<br />
A version of a dependency syntax for Wôpanâak is available [[Wôpanâak/Universal_Dependencies|here]].</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak&diff=4858Wôpanâak2017-04-25T02:38:22Z<p>Jmalin1: </p>
<hr />
<div>==External Resources==<br />
<br />
===Grammar===<br />
<br />
A grammar of the language published in 2000 by Jessie Little Doe Fermino (or Jessie Little Doe Baird) as an MIT thesis is available online [http://dspace.mit.edu/handle/1721.1/8740 here]. Most other resources I have found at this point are the sources used to develop this grammar. It is under copyright.<br />
<br />
Another [http://dspace.mit.edu/handle/1721.1/37422 thesis], by Nitana Hicks, lists morphemes used in Wôpanâak verbs along with rules for verb construction.<br />
<br />
Additionally, Volume 2 of ''Native Writings in Massachusett'' contains a "Grammatical Sketch" based on several centuries-older grammars.<br />
<br />
===Dictionary===<br />
<br />
The Natick Dictionary, published in 1903, appears to be a dictionary of at least one form of the language available online [https://archive.org/details/natickdictionar02trumgoog here]. It is a dictionary between English and the Natick dialect of Wôpanâak.<br />
<br />
===Corpora===<br />
<br />
The 1988 book ''Native Writings In Massachusett'' (available from McCabe and online), by Ives Goddard and Kathleen Bragdon, is the primary source of text for the language, containing many (possibly all) surviving texts written by native speakers of the language from the 1660s to the 1750s. The book contains many documents with transcriptions, annotations, and translations. The documents are in large part legal documents, including land transactions and wills, although documents with other purposes are in the collection as well. The book was published by the American Philosophical Society and is copyrighted.<br />
<br />
Additionally, a major source for the language is a translation of the Bible written by a non-native speaker (missionary John Eliot). Editions from [https://archive.org/details/mamussewunneetup00elio 1663] and [https://archive.org/details/mamussewunneetup02elio 1685] are available online, although they appear to be in image format without good transcription.<br />
<br />
==Developed Resources==<br />
<br />
===Keyboard Layouts===<br />
<br />
Preliminary [[Wôpanâak/Keyboard|keyboard layouts]] for Wôpanâak are available on [https://github.swarthmore.edu/jmalin1/ling073-wam-keyboard GitHub].<br />
<br />
[[Category:Wôpanâak]] [[Category:Sp17 ResourceDocumentation]]<br />
<br />
===Corpora===<br />
<br />
Basic corpora are available [https://github.swarthmore.edu/jmalin1/ling073-wam-corpus/graphs/contributors here] (still being constructed). The repository is private due to copyrighted sources for the corpora.<br />
<br />
===Morphological Analyzer===<br />
<br />
Some morphological processes are documented [[Wôpanâak/Grammar|here]].<br />
<br />
A preliminary morphological analyzer is under development [https://github.swarthmore.edu/jmalin1/ling073-wam here]. Currently it mainly analyzes nouns. More information can be found at [[Wôpanâak/Transducer]].<br />
<br />
===Disambiguation===<br />
<br />
A disambiguator is under development at [[Wôpanâak/Disambiguation]].<br />
<br />
===Khasi and Wôpanâak===<br />
<br />
A page containing all resources devoted to language pair Khasi and Wôpanâak can be found here: [[Khasi_and_Wôpanâak|Khasi and Wôpanâak]]<br />
<br />
===Contrastive Grammars===<br />
<br />
Differences in grammar between Khasi and Wôpanâak can be found at [[Khasi_and_Wôpanâak/ContrastiveGrammars|Contrastive Grammars]]<br />
<br />
==Universal Dependency==<br />
<br />
A version of a dependency syntax for Wôpanâak is available [[Wôpanâak/Universal_Dependencies|here]].</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak&diff=4782Khasi and Wôpanâak2017-04-24T04:17:59Z<p>Jmalin1: /* Improvements to Wôpanâak Transducer */</p>
<hr />
<div>Resources for machine translation between [[Khasi|Khasi]] and [[Wôpanâak|Wôpanâak]]<br />
* [[Khasi_and_Wôpanâak/ContrastiveGrammars|Contrastive Grammars]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Transfer]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Selection]]<br />
* [[Khasi_and_Wôpanâak/Structural_Transfer|Structural_Transfer]]<br />
[[Category:Sp17_TranslationPairs]]<br />
<br />
==Final Evaluation==<br />
<br />
===Improvements to Wôpanâak Transducer===<br />
*Updates to lexc<br />
**Removed {u} from first- and second- person suffixes, causing words such as "nutunantam" to analyze correctly. Additionally, {u} was added to third-person suffixes, allowing words like "nupuwak" to analyze correctly.<br />
**Fixed an overgeneration issue with intransitive verbs. Inanimate intransitive verbs can no longer receive first or second person morphology<br />
**Reworked transitive verbs with inanimate objects, splitting the inflection lexicon into two, one for absolute forms and one for objective forms<br />
**Added the "TI2" paradigm of verb stems to the transducer by sending them to a different lexicon before inflection. This has allowed the analyzer to handle words such as "ahtaw" ("have/own")<br />
**Added the verbalizing suffix -w to the noun pathway, allowing nouns with it to analyze as verbs (e.g. "sôtyumâw" - "he/she is sachem")<br />
**Added a number of words which should have been receiving the -m possession suffix to the correct category. Also updated the -m suffix to change to -um when following a consonant. This is hopefully correct, as it is consistent with phonology elsewhere in the language, but may require a further update as it is largely conjecture on my part.<br />
<br />
*Updates to twol<br />
**Updated rules for deletion of {m} and {w} to work after vowel archiphonemes as well as vowels.<br />
**Consolidated multiple rules sets that did the same thing to the same archiphoneme into single rules, clearing up rule conflicts in the compiler<br />
**Added a rule that deletes w from prefix of 3p possessed dependent nouns beginning with ȣ. Now the transducer correctly analyzes and generates words such as "ȣshah" instead of "wȣshah" for "his father".<br />
<br />
*Updates to twoc<br />
**Added tags and rules to prevent overgeneration of forms associated with the -w verbalizing suffix: nouns will not receive a 3p possession reading without the appropriate prefix, and the verb form will never receive the 3p possession prefix incorrectly.<br />
<br />
*Updates to dix<br />
**Added 100 words to dix file and lexc file<br />
<br />
*Evaluation<br />
**Annotated corpus precision and recall:<br />
***Precision : 92.6%<br />
***Recall: 80.6%<br />
**Number of tokenised words in corpus: 553<br />
**Corpus coverage: 50.45%<br />
**Number of stems in transducer: 167<br />
<br />
===Improvements to Khasi Transducer===<br />
*Updates to lexc<br />
**Completely restructured file and paths the transducer takes in order to make code much more readable, and fixed prefixation (I essentially rewrote most of the lexc file)<br />
**Added prefixed nominalization with 'kaba' (verbs and adverbs) and 'nym' and 'nong' (verbs)<br />
**Added progressives (prefixed) with 'nang' and 'iai'<br />
**Added future aspects with 'la'<br />
**Added past aspects with 'myn'<br />
**Added causality on verbs with 'pyn'<br />
**Added personal pronoun emphasis ('ma')<br />
**Added ungendered nouns and figured out how to work with them<br />
*Updates to disambiguator (x2)<br />
**Fixed 3rd and 4th disambiguator rules to be more specific to personal pronouns (If first word is a pronoun or an article, and next word is a noun/adj/verb, select article for the first word.)<br />
**All rules under "#Everything after this point was not written for the first assignment" in rlx file were written in the past but not as part of the disambiguator assignment; I found words that needed disambiguation so I fixed them as I found problems. There are three rules present that were not necessary for the first assignment, but that I completed previously.<br />
**For the last three rules necessary to (doubly) complete this portion, I wrote rules for selecting transitivity, differentiating between 'um' as a noun and 'um' as a negative personal pronoun, and getting a default for verbs that analyze twice (i.e. pyn+long vs pynlong, which are the same). See the rlx file for much more detail on all of the rules. <br />
*Updates to dix<br />
**Added 100 words to dix file and lexc file<br />
*Final Evaluation<br />
**My precision and recall test wasn't informative, as every time I saw an error (or undefined word) in my corpus, I added sections to the transducer the fixed those errors. I then regenerated until I was finished and didn't save the initial output before I started modifying the lexc file. The command outputs:<br />
***Totals: 357 tp, 0 fp, 0 tn, 0 fn<br />
***Precision: 100.00000%<br />
***Recall: 100.00000%<br />
***Because the files are identical.<br />
**Coverage over the large corpus is 55.67%<br />
**There are 53,444 words in my large corpus<br />
**There are 262 stems in the Khasi transducer<br />
===Evaluation of MT between languages===<br />
*Kha-Wam has 76.35% WER and PER<br />
*Kha-Wam has 45/82 stems translating correctly<br />
*Kha-Wam has 35.65% coverage and 1139 tokenized words <br />
*Wam-Kha has a 91.30% WER and PER<br />
*Wam-Kha has a 63/89 stems translating correctly<br />
*Wam-Kha has 25.51% coverage and 584 tokenized words</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak&diff=4781Khasi and Wôpanâak2017-04-24T04:17:35Z<p>Jmalin1: /* Improvements to Wôpanâak Transducer */</p>
<hr />
<div>Resources for machine translation between [[Khasi|Khasi]] and [[Wôpanâak|Wôpanâak]]<br />
* [[Khasi_and_Wôpanâak/ContrastiveGrammars|Contrastive Grammars]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Transfer]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Selection]]<br />
* [[Khasi_and_Wôpanâak/Structural_Transfer|Structural_Transfer]]<br />
[[Category:Sp17_TranslationPairs]]<br />
<br />
==Final Evaluation==<br />
<br />
===Improvements to Wôpanâak Transducer===<br />
*Updates to lexc<br />
**Removed {u} from first- and second- person suffixes, causing words such as "nutunantam" to analyze correctly. Additionally, {u} was added to third-person suffixes, allowing words like "nupuwak" to analyze correctly.<br />
**Fixed an overgeneration issue with intransitive verbs. Inanimate intransitive verbs can no longer receive first or second person morphology<br />
**Reworked transitive verbs with inanimate objects, splitting the inflection lexicon into two, one for absolute forms and one for objective forms<br />
**Added the "TI2" paradigm of verb stems to the transducer by sending them to a different lexicon before inflection. This has allowed the analyzer to handle words such as "ahtaw" ("have/own")<br />
**Added the verbalizing suffix -w to the noun pathway, allowing nouns with it to analyze as verbs (e.g. "sôtyumâw" - "he/she is sachem")<br />
**Added a number of words which should have been receiving the -m possession suffix to the correct category. Also updated the -m suffix to change to -um when following a consonant. This is hopefully correct, as it is consistent with phonology elsewhere in the language, but may require a further update as it is largely conjecture on my part.<br />
<br />
*Updates to twol<br />
**Updated rules for deletion of {m} and {w} to work after vowel archiphonemes as well as vowels.<br />
**Consolidated multiple rules sets that did the same thing to the same archiphoneme into single rules, clearing up rule conflicts in the compiler<br />
**Added a rule that deletes w from prefix of 3p possessed dependent nouns beginning with ȣ. Now the transducer correctly analyzes and generates words such as "ȣshah" instead of "wȣshah" for "his father".<br />
<br />
*Updates to twoc<br />
**Added tags and rules to prevent overgeneration of forms associated with the -w verbalizing suffix: nouns will not receive a 3p possession reading without the appropriate prefix, and the verb form will never receive the 3p possession prefix incorrectly.<br />
<br />
*Updates to dix<br />
**Added 100 words to dix file and lexc file<br />
<br />
*Evaluation<br />
**Annotated corpus precision and recall:<br />
***Precision : 92.6%<br />
***Recall: 80.6%<br />
**Number of tokenised words in corpus: 553<br />
**Corpus coverage: 50.45%<br />
**Number of stems in analyzer: 167<br />
<br />
===Improvements to Khasi Transducer===<br />
*Updates to lexc<br />
**Completely restructured file and paths the transducer takes in order to make code much more readable, and fixed prefixation (I essentially rewrote most of the lexc file)<br />
**Added prefixed nominalization with 'kaba' (verbs and adverbs) and 'nym' and 'nong' (verbs)<br />
**Added progressives (prefixed) with 'nang' and 'iai'<br />
**Added future aspects with 'la'<br />
**Added past aspects with 'myn'<br />
**Added causality on verbs with 'pyn'<br />
**Added personal pronoun emphasis ('ma')<br />
**Added ungendered nouns and figured out how to work with them<br />
*Updates to disambiguator (x2)<br />
**Fixed 3rd and 4th disambiguator rules to be more specific to personal pronouns (If first word is a pronoun or an article, and next word is a noun/adj/verb, select article for the first word.)<br />
**All rules under "#Everything after this point was not written for the first assignment" in rlx file were written in the past but not as part of the disambiguator assignment; I found words that needed disambiguation so I fixed them as I found problems. There are three rules present that were not necessary for the first assignment, but that I completed previously.<br />
**For the last three rules necessary to (doubly) complete this portion, I wrote rules for selecting transitivity, differentiating between 'um' as a noun and 'um' as a negative personal pronoun, and getting a default for verbs that analyze twice (i.e. pyn+long vs pynlong, which are the same). See the rlx file for much more detail on all of the rules. <br />
*Updates to dix<br />
**Added 100 words to dix file and lexc file<br />
*Final Evaluation<br />
**My precision and recall test wasn't informative, as every time I saw an error (or undefined word) in my corpus, I added sections to the transducer the fixed those errors. I then regenerated until I was finished and didn't save the initial output before I started modifying the lexc file. The command outputs:<br />
***Totals: 357 tp, 0 fp, 0 tn, 0 fn<br />
***Precision: 100.00000%<br />
***Recall: 100.00000%<br />
***Because the files are identical.<br />
**Coverage over the large corpus is 55.67%<br />
**There are 53,444 words in my large corpus<br />
**There are 262 stems in the Khasi transducer<br />
===Evaluation of MT between languages===<br />
*Kha-Wam has 76.35% WER and PER<br />
*Kha-Wam has 45/82 stems translating correctly<br />
*Kha-Wam has 35.65% coverage and 1139 tokenized words <br />
*Wam-Kha has a 91.30% WER and PER<br />
*Wam-Kha has a 63/89 stems translating correctly<br />
*Wam-Kha has 25.51% coverage and 584 tokenized words</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak&diff=4779Khasi and Wôpanâak2017-04-24T03:56:03Z<p>Jmalin1: /* Improvements to Wôpanâak Transducer */</p>
<hr />
<div>Resources for machine translation between [[Khasi|Khasi]] and [[Wôpanâak|Wôpanâak]]<br />
* [[Khasi_and_Wôpanâak/ContrastiveGrammars|Contrastive Grammars]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Transfer]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Selection]]<br />
* [[Khasi_and_Wôpanâak/Structural_Transfer|Structural_Transfer]]<br />
[[Category:Sp17_TranslationPairs]]<br />
<br />
==Final Evaluation==<br />
<br />
===Improvements to Wôpanâak Transducer===<br />
*Updates to lexc<br />
**Removed {u} from first- and second- person suffixes, causing words such as "nutunantam" to analyze correctly. Additionally, {u} was added to third-person suffixes, allowing words like "nupuwak" to analyze correctly.<br />
**Fixed an overgeneration issue with intransitive verbs. Inanimate intransitive verbs can no longer receive first or second person morphology<br />
**Reworked transitive verbs with inanimate objects, splitting the inflection lexicon into two, one for absolute forms and one for objective forms<br />
**Added the "TI2" paradigm of verb stems to the transducer by sending them to a different lexicon before inflection. This has allowed the analyzer to handle words such as "ahtaw" ("have/own")<br />
**Added the verbalizing suffix -w to the noun pathway, allowing nouns with it to analyze as verbs (e.g. "sôtyumâw" - "he/she is sachem")<br />
**Added a number of words which should have been receiving the -m possession suffix to the correct category. Also updated the -m suffix to change to -um when following a consonant. This is hopefully correct, as it is consistent with phonology elsewhere in the language, but may require a further update as it is largely conjecture on my part.<br />
<br />
*Updates to twol<br />
**Updated rules for deletion of {m} and {w} to work after vowel archiphonemes as well as vowels.<br />
**Consolidated multiple rules sets that did the same thing to the same archiphoneme into single rules, clearing up rule conflicts in the compiler<br />
**Added a rule that deletes w from prefix of 3p possessed dependent nouns beginning with ȣ. Now the transducer correctly analyzes and generates words such as "ȣshah" instead of "wȣshah" for "his father".<br />
<br />
*Updates to twoc<br />
**Added tags and rules to prevent overgeneration of forms associated with the -w verbalizing suffix: nouns will not receive a 3p possession reading without the appropriate prefix, and the verb form will never receive the 3p possession prefix incorrectly.<br />
<br />
*Updates to dix<br />
**Added 100 words to dix file and lexc file<br />
<br />
===Improvements to Khasi Transducer===<br />
*Updates to lexc<br />
**Completely restructured file and paths the transducer takes in order to make code much more readable, and fixed prefixation (I essentially rewrote most of the lexc file)<br />
**Added prefixed nominalization with 'kaba' (verbs and adverbs) and 'nym' and 'nong' (verbs)<br />
**Added progressives (prefixed) with 'nang' and 'iai'<br />
**Added future aspects with 'la'<br />
**Added past aspects with 'myn'<br />
**Added causality on verbs with 'pyn'<br />
**Added personal pronoun emphasis ('ma')<br />
**Added ungendered nouns and figured out how to work with them<br />
*Updates to disambiguator (x2)<br />
**Fixed 3rd and 4th disambiguator rules to be more specific to personal pronouns (If first word is a pronoun or an article, and next word is a noun/adj/verb, select article for the first word.)<br />
**All rules under "#Everything after this point was not written for the first assignment" in rlx file were written in the past but not as part of the disambiguator assignment; I found words that needed disambiguation so I fixed them as I found problems. There are three rules present that were not necessary for the first assignment, but that I completed previously.<br />
**For the last three rules necessary to (doubly) complete this portion, I wrote rules for selecting transitivity, differentiating between 'um' as a noun and 'um' as a negative personal pronoun, and getting a default for verbs that analyze twice (i.e. pyn+long vs pynlong, which are the same). See the rlx file for much more detail on all of the rules. <br />
*Updates to dix<br />
**Added 100 words to dix file and lexc file<br />
*Final Evaluation<br />
**My precision and recall test wasn't informative, as every time I saw an error (or undefined word) in my corpus, I added sections to the transducer the fixed those errors. I then regenerated until I was finished and didn't save the initial output before I started modifying the lexc file. The command outputs:<br />
***Totals: 357 tp, 0 fp, 0 tn, 0 fn<br />
***Precision: 100.00000%<br />
***Recall: 100.00000%<br />
***Because the files are identical.<br />
**Coverage over the large corpus is 55.67%<br />
**There are 53,444 words in my large corpus<br />
**There are 262 stems in the Khasi transducer<br />
===Evaluation of MT between languages===<br />
*Kha-Wam has 76.35% WER and PER<br />
*Kha-Wam has 45/82 stems translating correctly<br />
*Kha-Wam has 35.65% coverage and 1139 tokenized words <br />
*Wam-Kha has a 91.30% WER and PER<br />
*Wam-Kha has a 63/89 stems translating correctly<br />
*Wam-Kha has 25.51% coverage and 584 tokenized words</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak/Disambiguation&diff=4777Wôpanâak/Disambiguation2017-04-24T03:55:08Z<p>Jmalin1: </p>
<hr />
<div>The disambiguator is available as part of the transducer [https://github.swarthmore.edu/jmalin1/ling073-wam here].<br />
<br />
==Initial Evaluation of Ambiguity==<br />
Initial ambiguity is calculated at 1.026. However, this analysis may be impacted by tokenisation errors.<br />
<br />
==Example sentences==<br />
^yȣ/'''<prn><dem><nn><prx><sg>'''/<det><dem><nn><prx><sg>$ ^mata/mata<adv>$ ^numâkȣwun/mâkȣ<vblex><tv><p1><neg><obj>$<br />
<br />
"This I do not sell"<br />
<br />
^yȣ/<prn><dem><nn><prx><sg>/'''<det><dem><nn><prx><sg>'''$ ^umâkȣôk/mâkȣôk<n><nn><px3sg><sg>$ ^pômpakesay/pômpakes<n><aa><sg><abs>$<br />
<br />
"This the conveyance of the late Pompakes"<br />
<br />
It is somewhat unclear which tags are correct here: as a statement of identity, it's possible that yȣ should in fact be analyzed as a pronoun, which means that the disambiguation rule would produce the wrong result in this case.<br />
<br />
==Final evaluation of ambiguity==<br />
Final ambiguity is calculated at 1.011.<br />
<br />
[[Category:Wôpanâak]][[Category:Sp17_Disambiguation]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak/Disambiguation&diff=4775Wôpanâak/Disambiguation2017-04-24T02:03:23Z<p>Jmalin1: </p>
<hr />
<div>==Initial Evaluation of Ambiguity==<br />
Initial ambiguity is calculated at 1.026. However, this analysis may be impacted by tokenisation errors.<br />
<br />
==Example sentences==<br />
^yȣ/'''<prn><dem><nn><prx><sg>'''/<det><dem><nn><prx><sg>$ ^mata/mata<adv>$ ^numâkȣwun/mâkȣ<vblex><tv><p1><neg><obj>$<br />
<br />
"This I do not sell"<br />
<br />
^yȣ/<prn><dem><nn><prx><sg>/'''<det><dem><nn><prx><sg>'''$ ^umâkȣôk/mâkȣôk<n><nn><px3sg><sg>$ ^pômpakesay/pômpakes<n><aa><sg><abs>$<br />
<br />
"This the conveyance of the late Pompakes"<br />
<br />
It is somewhat unclear which tags are correct here: as a statement of identity, it's possible that yȣ should in fact be analyzed as a pronoun, which means that the disambiguation rule would produce the wrong result in this case.<br />
<br />
==Final evaluation of ambiguity==<br />
Final ambiguity is calculated at 1.011.<br />
<br />
[[Category:Wôpanâak]][[Category:Sp17_Disambiguation]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak&diff=4647Khasi and Wôpanâak2017-04-19T05:44:07Z<p>Jmalin1: /* Improvements to Wôpanâak Transducer */</p>
<hr />
<div>Resources for machine translation between [[Khasi|Khasi]] and [[Wôpanâak|Wôpanâak]]<br />
* [[Khasi_and_Wôpanâak/ContrastiveGrammars|Contrastive Grammars]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Transfer]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Selection]]<br />
* [[Khasi_and_Wôpanâak/Structural_Transfer|Structural_Transfer]]<br />
[[Category:Sp17_TranslationPairs]]<br />
<br />
==Final Evaluation==<br />
<br />
===Improvements to Wôpanâak Transducer===<br />
*Updates to lexc<br />
**Removed {u} from first- and second- person suffixes, causing words such as "nutunantam" to analyze correctly. Additionally, {u} was added to third-person suffixes, allowing words like "nupuwak" to analyze correctly.<br />
**Fixed an overgeneration issue with intransitive verbs. Inanimate intransitive verbs can no longer receive first or second person morphology<br />
**Reworked transitive verbs with inanimate objects, splitting the inflection lexicon into two, one for absolute forms and one for objective forms<br />
**Added the "TI2" paradigm of verb stems to the transducer by sending them to a different lexicon before inflection. This has allowed the analyzer to handle words such as "ahtaw" ("have/own")<br />
**Added the verbalizing suffix -w to the noun pathway, allowing nouns with it to analyze as verbs (e.g. "sôtyumâw" - "he/she is sachem")<br />
**Added a number of words which should have been receiving the -m possession suffix to the correct category. Also updated the -m suffix to change to -um when following a consonant. This is hopefully correct, as it is consistent with phonology elsewhere in the language, but may require a further update as it is largely conjecture on my part.<br />
<br />
*Updates to twol<br />
**Updated rules for deletion of {m} and {w} to work after vowel archiphonemes as well as vowels.<br />
**Consolidated multiple rules sets that did the same thing to the same archiphoneme into single rules, clearing up rule conflicts in the compiler<br />
**Added a rule that deletes w from prefix of 3p possessed dependent nouns beginning with ȣ. Now the transducer correctly analyzes and generates words such as "ȣshah" instead of "wȣshah" for "his father".<br />
<br />
*Updates to twoc<br />
**Added tags and rules to prevent overgeneration of forms associated with the -w verbalizing suffix: nouns will not receive a 3p possession reading without the appropriate prefix, and the verb form will never receive the 3p possession prefix incorrectly.<br />
<br />
===Improvements to Khasi Transducer===<br />
*Updates to lexc<br />
**Completely restructured file and paths the transducer takes in order to make code much more readable, and fixed prefixation (I essentially rewrote most of the lexc file)<br />
**Added prefixed nominalization with 'kaba' (verbs and adverbs) and 'nym' and 'nong' (verbs)<br />
**Added progressives (prefixed) with 'nang' and 'iai'<br />
**Added future aspects with 'la'<br />
**Added past aspects with 'myn'<br />
**Added causality on verbs with 'pyn'<br />
**Added personal pronoun emphasis ('ma')<br />
**Added ungendered nouns and figured out how to work with them<br />
*Updates to disambiguator<br />
**Fixed 4th disambiguator rule to work with verbs (If first word is a pronoun or an article, and next word is a noun/adj/verb, select article for the first word.)<br />
**All rules under #----------------------------- in rlx file were written in the past but not as part of the disambiguator assignment; I found words that needed disambiguation so I fixed them as I found problems. There are three rules present that were not necessary for the first assignment.</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak&diff=4646Khasi and Wôpanâak2017-04-19T05:42:45Z<p>Jmalin1: /* Improvements to Wôpanâak Transducer */</p>
<hr />
<div>Resources for machine translation between [[Khasi|Khasi]] and [[Wôpanâak|Wôpanâak]]<br />
* [[Khasi_and_Wôpanâak/ContrastiveGrammars|Contrastive Grammars]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Transfer]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Selection]]<br />
* [[Khasi_and_Wôpanâak/Structural_Transfer|Structural_Transfer]]<br />
[[Category:Sp17_TranslationPairs]]<br />
<br />
==Final Evaluation==<br />
<br />
===Improvements to Wôpanâak Transducer===<br />
*Updates to lexc<br />
**Removed {u} from first- and second- person suffixes, causing words such as "nutunantam" to analyze correctly. Additionally, {u} was added to third-person suffixes, allowing words like "nupuwak" to analyze correctly.<br />
**Fixed an overgeneration issue with intransitive verbs. Inanimate intransitive verbs can no longer receive first or second person morphology<br />
**Reworked transitive verbs with inanimate objects, splitting the inflection lexicon into two, one for absolute forms and one for objective forms<br />
**Added the "TI2" paradigm of verb stems to the transducer by sending them to a different lexicon before inflection. This has allowed the analyzer to handle words such as "ahtaw" ("have/own")<br />
**Added the verbalizing suffix -w to the noun pathway, allowing nouns with it to analyze as verbs (e.g. "sôtyumâw" - "he/she is sachem")<br />
**Added a number of words which should have been receiving the -m possession suffix to the correct category. Also updated the -m suffix to change to -um when following a consonant. This is hopefully correct, as it is consistent with phonology elsewhere in the language, but may require a further update as it is largely conjecture on my part.<br />
<br />
*Updates to twol<br />
**Updated rules for deletion of {m} and {w} to work after vowel archiphonemes as well as vowels.<br />
**Consolidated multiple rules sets that did the same thing to the same archiphoneme into single rules, clearing up rule conflicts in the compiler<br />
**Added a rule that deletes w from prefix of 3p possessed dependent nouns beginning with ȣ. Now the transducer correctly analyzes and generates words such as "ȣshah" instead of "wȣshah" for "his father".<br />
<br />
*Updates to twoc<br />
**Added tags and rules to prevent overgeneration of forms associated with the -w verbalizing suffix: nouns will not receive an alternate 3p possession reading without the appropriate prefix, and the verb form will never receive the 3p possession prefix incorrectly.<br />
<br />
===Improvements to Khasi Transducer===<br />
*Updates to lexc<br />
**Completely restructured file and paths the transducer takes in order to make code much more readable, and fixed prefixation (I essentially rewrote most of the lexc file)<br />
**Added prefixed nominalization with 'kaba' (verbs and adverbs) and 'nym' and 'nong' (verbs)<br />
**Added progressives (prefixed) with 'nang' and 'iai'<br />
**Added future aspects with 'la'<br />
**Added past aspects with 'myn'<br />
**Added causality on verbs with 'pyn'<br />
**Added personal pronoun emphasis ('ma')<br />
**Added ungendered nouns and figured out how to work with them<br />
*Updates to disambiguator<br />
**Fixed 4th disambiguator rule to work with verbs (If first word is a pronoun or an article, and next word is a noun/adj/verb, select article for the first word.)<br />
**All rules under #----------------------------- in rlx file were written in the past but not as part of the disambiguator assignment; I found words that needed disambiguation so I fixed them as I found problems. There are three rules present that were not necessary for the first assignment.</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak&diff=4645Khasi and Wôpanâak2017-04-19T05:38:24Z<p>Jmalin1: /* Improvements to Wôpanâak Transducer */</p>
<hr />
<div>Resources for machine translation between [[Khasi|Khasi]] and [[Wôpanâak|Wôpanâak]]<br />
* [[Khasi_and_Wôpanâak/ContrastiveGrammars|Contrastive Grammars]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Transfer]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Selection]]<br />
* [[Khasi_and_Wôpanâak/Structural_Transfer|Structural_Transfer]]<br />
[[Category:Sp17_TranslationPairs]]<br />
<br />
==Final Evaluation==<br />
<br />
===Improvements to Wôpanâak Transducer===<br />
*Updates to lexc<br />
**Removed {u} from first- and second- person suffixes, causing words such as "nutunantam" to analyze correctly. Additionally, {u} was added to third-person suffixes, allowing words like "nupuwak" to analyze correctly.<br />
**Fixed an overgeneration issue with intransitive verbs. Inanimate intransitive verbs can no longer receive first or second person morphology<br />
**Reworked transitive verbs with inanimate objects, splitting the inflection lexicon into two, one for absolute forms and one for objective forms<br />
**Added the "TI2" paradigm of verb stems to the transducer by sending them to a different lexicon before inflection. This has allowed the analyzer to handle words such as "ahtaw" ("have/own")<br />
**Added the verbalizing suffix -w to the noun pathway, allowing nouns with it to analyze as verbs (e.g. "sôtyumâw" - "he/she is sachem")<br />
<br />
*Updates to twol<br />
**Updated rules for deletion of {m} and {w} to work after vowel archiphonemes as well as vowels.<br />
**Consolidated multiple rules sets that did the same thing to the same archiphoneme into single rules, clearing up rule conflicts in the compiler<br />
**Added a rule that deletes w from prefix of 3p possessed dependent nouns beginning with ȣ. Now the transducer correctly analyzes and generates words such as "ȣshah" instead of "wȣshah" for "his father".<br />
<br />
*Updates to twoc<br />
**Added tags and rules to prevent overgeneration of forms associated with the -w verbalizing suffix: nouns will not receive an alternate 3p possession reading without the appropriate prefix, and the verb form will never receive the 3p possession prefix incorrectly.<br />
<br />
===Improvements to Khasi Transducer===<br />
*Updates to lexc<br />
**Completely restructured file and paths the transducer takes in order to make code much more readable, and fixed prefixation (I essentially rewrote most of the lexc file)<br />
**Added prefixed nominalization with 'kaba' (verbs and adverbs) and 'nym' and 'nong' (verbs)<br />
**Added progressives (prefixed) with 'nang' and 'iai'<br />
**Added future aspects with 'la'<br />
**Added past aspects with 'myn'<br />
**Added causality on verbs with 'pyn'<br />
**Added personal pronoun emphasis ('ma')<br />
**Added ungendered nouns and figured out how to work with them<br />
*Updates to disambiguator<br />
**Fixed 4th disambiguator rule to work with verbs (If first word is a pronoun or an article, and next word is a noun/adj/verb, select article for the first word.)<br />
**All rules under #----------------------------- in rlx file were written in the past but not as part of the disambiguator assignment; I found words that needed disambiguation so I fixed them as I found problems. There are three rules present that were not necessary for the first assignment.</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak&diff=4644Khasi and Wôpanâak2017-04-19T05:26:31Z<p>Jmalin1: </p>
<hr />
<div>Resources for machine translation between [[Khasi|Khasi]] and [[Wôpanâak|Wôpanâak]]<br />
* [[Khasi_and_Wôpanâak/ContrastiveGrammars|Contrastive Grammars]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Transfer]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Selection]]<br />
* [[Khasi_and_Wôpanâak/Structural_Transfer|Structural_Transfer]]<br />
[[Category:Sp17_TranslationPairs]]<br />
<br />
==Final Evaluation==<br />
<br />
===Improvements to Wôpanâak Transducer===<br />
*Updates to lexc<br />
**Removed {u} from first- and second- person suffixes, causing words such as "nutunantam" to analyze correctly. Additionally, {u} was added to third-person suffixes, allowing words like "nupuwak" to analyze correctly.<br />
**Fixed an overgeneration issue with intransitive verbs. Inanimate intransitive verbs can no longer receive first or second person morphology<br />
**Reworked transitive verbs with inanimate objects, splitting the inflection lexicon into two, one for absolute forms and one for objective forms<br />
**Added the "TI2" paradigm of verb stems to the transducer by sending them to a different lexicon before inflection. This has allowed the analyzer to handle words such as "ahtaw" ("have/own")<br />
**Added the verbalizing suffix -w to the noun pathway, allowing nouns with it to analyze as verbs (e.g. "sôtyumâw" - "he/she is sachem")<br />
<br />
*Updates to twol<br />
**Updated rules for deletion of {m} and {w} to work after vowel archiphonemes as well as vowels.<br />
**Consolidated multiple rules sets that did the same thing to the same archiphoneme into single rules, clearing up rule conflicts in the compiler<br />
**Added a rule that deletes w from prefix of 3p possessed dependent nouns beginning with ȣ. Now the transducer correctly analyzes and generates words such as "ȣshah" instead of "wȣshah" for "his father".<br />
<br />
===Improvements to Khasi Transducer===<br />
*Updates to lexc<br />
**Completely restructured file and paths the transducer takes in order to make code much more readable, and fixed prefixation (I essentially rewrote most of the lexc file)<br />
**Added prefixed nominalization with 'kaba' (verbs and adverbs) and 'nym' and 'nong' (verbs)<br />
**Added progressives (prefixed) with 'nang' and 'iai'<br />
**Added future aspects with 'la'<br />
**Added past aspects with 'myn'<br />
**Added causality on verbs with 'pyn'<br />
**Added personal pronoun emphasis ('ma')<br />
**Added ungendered nouns and figured out how to work with them<br />
*Updates to disambiguator<br />
**Fixed 4th disambiguator rule to work with verbs (If first word is a pronoun or an article, and next word is a noun/adj/verb, select article for the first word.)<br />
**All rules under #----------------------------- in rlx file were written in the past but not as part of the disambiguator assignment; I found words that needed disambiguation so I fixed them as I found problems. There are three rules present that were not necessary for the first assignment.</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak&diff=4643Khasi and Wôpanâak2017-04-19T05:24:45Z<p>Jmalin1: /* Improvements to Wôpanâak Transducer */</p>
<hr />
<div>Resources for machine translation between [[Khasi|Khasi]] and [[Wôpanâak|Wôpanâak]]<br />
* [[Khasi_and_Wôpanâak/ContrastiveGrammars|Contrastive Grammars]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Transfer]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Selection]]<br />
* [[Khasi_and_Wôpanâak/Structural_Transfer|Structural_Transfer]]<br />
[[Category:Sp17_TranslationPairs]]<br />
<br />
==Final Evaluation==<br />
<br />
===Improvements to Wôpanâak Transducer===<br />
*Updates to lexc<br />
**Removed {u} from first- and second- person suffixes, causing words such as "nutunantam" to analyze correctly. Additionally, {u} was added to third-person suffixes, allowing words like "nupuwak" to analyze correctly.<br />
**Fixed an overgeneration issue with intransitive verbs. Inanimate intransitive verbs can no longer receive first or second person morphology<br />
**Reworked transitive verbs with inanimate objects, splitting the inflection lexicon into two, one for absolute forms and one for objective forms<br />
**Added the "TI2" paradigm of verb stems to the transducer by sending them to a different lexicon before inflection. This has allowed the analyzer to handle words such as "ahtaw" ("have/own")<br />
**Added the verbalizing suffix -w to the noun pathway, allowing nouns with it to analyze as verbs (e.g. "sôtyumâw" - "he/she is sachem")<br />
<br />
*Updates to twol<br />
**Updated rules for deletion of {m} and {w} to work after vowel archiphonemes as well as vowels.<br />
**Consolidated multiple rules sets that did the same thing to the same archiphoneme into single rules, clearing up rule conflicts in the compiler<br />
**Added a rule that deletes w from prefix of 3p possessed dependent nouns beginning with ȣ.<br />
<br />
===Improvements to Khasi Transducer===<br />
*Updates to lexc<br />
**Completely restructured file and paths the transducer takes in order to make code much more readable, and fixed prefixation (I essentially rewrote most of the lexc file)<br />
**Added prefixed nominalization with 'kaba' (verbs and adverbs) and 'nym' and 'nong' (verbs)<br />
**Added progressives (prefixed) with 'nang' and 'iai'<br />
**Added future aspects with 'la'<br />
**Added past aspects with 'myn'<br />
**Added causality on verbs with 'pyn'<br />
**Added personal pronoun emphasis ('ma')<br />
**Added ungendered nouns and figured out how to work with them<br />
*Updates to disambiguator<br />
**Fixed 4th disambiguator rule to work with verbs (If first word is a pronoun or an article, and next word is a noun/adj/verb, select article for the first word.)<br />
**All rules under #----------------------------- in rlx file were written in the past but not as part of the disambiguator assignment; I found words that needed disambiguation so I fixed them as I found problems. There are three rules present that were not necessary for the first assignment.</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak&diff=4604Khasi and Wôpanâak2017-04-18T06:05:27Z<p>Jmalin1: /* =Improvements to Wôpanâak Transducer */</p>
<hr />
<div>Resources for machine translation between [[Khasi|Khasi]] and [[Wôpanâak|Wôpanâak]]<br />
* [[Khasi_and_Wôpanâak/ContrastiveGrammars|Contrastive Grammars]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Transfer]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Selection]]<br />
* [[Khasi_and_Wôpanâak/Structural_Transfer|Structural_Transfer]]<br />
[[Category:Sp17_TranslationPairs]]<br />
<br />
==Final Evaluation==<br />
<br />
===Improvements to Wôpanâak Transducer===<br />
*Updates to lexc<br />
**Removed {u} from first- and second- person suffixes, causing words such as "nutunantam" to analyze correctly. Additionally, {u} was added to third-person suffixes, allowing words like "nupuwak" to analyze correctly.<br />
**Fixed an overgeneration issue with intransitive verbs. Inanimate intransitive verbs can no longer receive first or second person morphology<br />
**Reworked transitive verbs with inanimate objects, splitting the inflection lexicon into two, one for absolute forms and one for objective forms<br />
**Added the "TI2" paradigm of verb stems to the transducer by sending them to a different lexicon before inflection. This has allowed the analyzer to handle words such as "ahtaw" ("have/own")<br />
<br />
*Updates to twol<br />
**Updated rules for deletion of {m} and {w} to work after vowel archiphonemes as well as vowels.<br />
**Consolidated multiple rules sets that did the same thing to the same archiphoneme into single rules, clearing up rule conflicts in the compiler (in progress)</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak&diff=4603Khasi and Wôpanâak2017-04-18T06:05:04Z<p>Jmalin1: </p>
<hr />
<div>Resources for machine translation between [[Khasi|Khasi]] and [[Wôpanâak|Wôpanâak]]<br />
* [[Khasi_and_Wôpanâak/ContrastiveGrammars|Contrastive Grammars]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Transfer]]<br />
* [[Khasi_and_Wôpanâak/Lexical_Selection|Lexical_Selection]]<br />
* [[Khasi_and_Wôpanâak/Structural_Transfer|Structural_Transfer]]<br />
[[Category:Sp17_TranslationPairs]]<br />
<br />
==Final Evaluation==<br />
<br />
===Improvements to Wôpanâak Transducer==<br />
*Updates to lexc<br />
**Removed {u} from first- and second- person suffixes, causing words such as "nutunantam" to analyze correctly. Additionally, {u} was added to third-person suffixes, allowing words like "nupuwak" to analyze correctly.<br />
**Fixed an overgeneration issue with intransitive verbs. Inanimate intransitive verbs can no longer receive first or second person morphology<br />
**Reworked transitive verbs with inanimate objects, splitting the inflection lexicon into two, one for absolute forms and one for objective forms<br />
**Added the "TI2" paradigm of verb stems to the transducer by sending them to a different lexicon before inflection. This has allowed the analyzer to handle words such as "ahtaw" ("have/own")<br />
<br />
*Updates to twol<br />
**Updated rules for deletion of {m} and {w} to work after vowel archiphonemes as well as vowels.<br />
**Consolidated multiple rules sets that did the same thing to the same archiphoneme into single rules, clearing up rule conflicts in the compiler (in progress)</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak/Disambiguation&diff=4553Wôpanâak/Disambiguation2017-04-17T02:07:22Z<p>Jmalin1: /* Example sentences */</p>
<hr />
<div>==Initial Evaluation of Ambiguity==<br />
Initial ambiguity is calculated at 1.026. However, this analysis may be impacted by tokenisation errors.<br />
<br />
==Example sentences==<br />
^yȣ/'''<prn><dem><nn><prx><sg>'''/<det><dem><nn><prx><sg>$ ^mata/mata<adv>$ ^numâkȣwun/mâkȣ<vblex><tv><p1><neg><obj>$<br />
<br />
"This I do not sell"<br />
<br />
^yȣ/<prn><dem><nn><prx><sg>/'''<det><dem><nn><prx><sg>'''$ ^umâkȣôk/mâkȣôk<n><nn><px3sg><sg>$ ^pômpakesay/pômpakes<n><aa><sg><abs>$<br />
<br />
"This the conveyance of the late Pompakes"<br />
<br />
It is somewhat unclear which tags are correct here: as a statement of identity, it's possible that yȣ should in fact be analyzed as a pronoun, which means that the disambiguation rule would produce the wrong result in this case.<br />
<br />
<br />
[[Category:Wôpanâak]][[Category:Sp17_Disambiguation]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak/Disambiguation&diff=4549Wôpanâak/Disambiguation2017-04-17T01:16:20Z<p>Jmalin1: </p>
<hr />
<div>==Initial Evaluation of Ambiguity==<br />
Initial ambiguity is calculated at 1.026. However, this analysis may be impacted by tokenisation errors.<br />
<br />
==Example sentences==<br />
^yȣ/'''<prn><dem><nn><prx><sg>'''/<det><dem><nn><prx><sg>$ ^mata/mata<adv>$ ^numâkȣwun/mâkȣ<vblex><tv><p1><neg><obj>$<br />
<br />
"This I do not sell"<br />
<br />
^yȣ/<prn><dem><nn><prx><sg>/'''<det><dem><nn><prx><sg>'''$ ^umâkȣôk/mâkȣôk<n><nn><px3sg><sg>$ ^pômpakesay/pômpakes<n><aa><sg><abs>$<br />
<br />
"This the conveyance of the late Pompakes"<br />
<br />
It is somewhat unclear which translation is correct here.<br />
<br />
<br />
[[Category:Wôpanâak]][[Category:Sp17_Disambiguation]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak/Disambiguation&diff=4548Wôpanâak/Disambiguation2017-04-17T01:12:05Z<p>Jmalin1: </p>
<hr />
<div>==Initial Evaluation of Ambiguity==<br />
Initial ambiguity is calculated at 1.026. However, this analysis may be impacted by tokenisation errors.<br />
<br />
==Example sentences==<br />
^yȣ/'''<prn><dem><nn><prx><sg>'''/<det><dem><nn><prx><sg>$ ^mata/mata<adv>$ ^numâkȣwun/mâkȣ<vblex><tv><p1><neg><obj>$<br />
"This I do not sell"<br />
<br />
^yȣ/<prn><dem><nn><prx><sg>/'''<det><dem><nn><prx><sg>'''$ ^umâkȣôk/mâkȣôk<n><nn><px3sg><sg>$ ^pômpakesay/pômpakes<n><aa><sg><abs>$<br />
"This the conveyance of the late Pompakes"<br />
It is somewhat unclear which translation is correct here.<br />
<br />
<br />
[[Category:Wôpanâak]][[Category:Sp17_Disambiguation]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak/Lexical_Selection&diff=4443Khasi and Wôpanâak/Lexical Selection2017-04-11T00:25:09Z<p>Jmalin1: </p>
<hr />
<div>===wam-->kha===<br />
*meequn "feather, pen" => sner "feather", khulom "pen", khulom-sner "feather pen"<br />
**if bird, arrow near meequn, select sner<br />
**if write, paper, book near meedun, select khulom<br />
*keesuk "sky, a day" => sngi "day, sun", bneng "sky"<br />
**if color-word near keesuk, choose bneng<br />
**if countable-word, plural near keesuk, choose sngi<br />
*nupâus "sun, month" => sngi "day, sun", bnai "moon, month"<br />
** if rise, set near nupâus, select sngi !!!Does Wôpanâak have words for rise and set?<br />
** else default to bnai<br />
<br />
====Examples====<br />
*In "monatash keesukash" ("many days"), "keesuk" should be translated to "sngi"<br />
*In "ayum God keesuk kah ahkee" ("God creates the heavens and the earth"), "keesuk" should be translated to "bneng".<br />
<br />
===kha-->wam===<br />
* kjat "foot/leg" => -ukhôt "leg", -seet "foot"<br />
** default to -seet<br />
* sngi "day, sun" => keesuk "sky, day", nupâus "sun, month"<br />
** if what, which, long, short, PLURAL near sngi, select keesuk<br />
** if rise, set (sep) near sngi, select nupâus<br />
* OLD - for reference for later: ngi "we (1st person plural pronoun)" => neenawun "we (exclusive)", keenawun "we (inclusive)" This is reflective of a deeper grammatical ambiguity with regards to first person plural forms throughout the languages.<br />
<br />
====Examples====<br />
*In "bad ia ki sngi" ("and days"), "sngi" should be translated to "keesuk"<br />
*In "ka sngi ka la mih la slem" ("the sun was up long ago"), "sngi" should be translated to "nupâus"<br />
<br />
[[Category:Sp17_LexicalSelection]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak/Structural_Transfer&diff=4367Khasi and Wôpanâak/Structural Transfer2017-04-07T03:27:15Z<p>Jmalin1: </p>
<hr />
<div>==Pre-Evaluation==<br />
===WER test===<br />
*13/17 correct<br />
===PER test===<br />
*23.53% unknown words<br />
===Coverage tests===<br />
Khasi<br />
*<br />
Wôpanâak<br />
*<br />
==Example Sentences==<br />
u dngiem u phet "The bear runs"<br />
<br />
* tagger: ^article<art><m><sg>$ ^dngiem<n><m>$ ^article<art><m><sg>$ ^phet<vblex><iv>$<br />
* biltrans: ^article<art><m><sg>/$ ^dngiem<n><m>/masq<n><aa>$ ^article<art><m><sg>/$ ^phet<vblex><iv>/qaqee<vblex><iv>$<br />
* chunker: ^n<SN><aa><sg>{^masq<n><2><3>$}$ ^default<default>{^qaqee<vblex><iv>$}$<br />
* postchunk: ^masq<n><aa><sg>$ ^qaqee<vblex><iv>$<br />
* wam-kha: masq #qaqee (There is not a yet a structural transfer rule for subject agreement on verbs)<br />
<br />
masqak qaqeewak "The bears run"<br />
<br />
* tagger: ^masq<n><aa><pl>$ ^qaqee<vblex><iv><aa><p3><pl>$<br />
* biltrans: ^masq<n><aa><pl>/dngiem<n><m><pl>$ ^qaqee<vblex><iv><aa><p3><pl>/phet<vblex><iv><aa><p3><pl>$<br />
* chunker: ^art<SA><m><pl>{^article<art><3>$}$ ^n<SN><m><pl>{^dngiem<n><2>$}$ ^default<default>{^phet<vblex><iv><aa><p3><pl>$}$<br />
* postchunk: ^article<art><pl>$ ^dngiem<n><m>$ ^phet<vblex><iv><aa><p3><pl>$<br />
* wam-kha: ki dngiem #phet (Verb agreement rules are not implemented in this direction either)<br />
<br />
==Post-Evaluation==<br />
<br />
[[Category:Sp17_StructuralTransfer]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak/Structural_Transfer&diff=4366Khasi and Wôpanâak/Structural Transfer2017-04-07T03:26:25Z<p>Jmalin1: </p>
<hr />
<div>==Pre-Evaluation==<br />
===WER test===<br />
*13/17 correct<br />
===PER test===<br />
*23.53% unknown words<br />
===Coverage tests===<br />
Khasi<br />
*<br />
Wôpanâak<br />
*<br />
==Example Sentences==<br />
u dngiem u phet "The bear runs"<br />
<br />
* tagger: ^article<art><m><sg>$ ^dngiem<n><m>$ ^article<art><m><sg>$ ^phet<vblex><iv>$<br />
* biltrans: ^article<art><m><sg>/$ ^dngiem<n><m>/masq<n><aa>$ ^article<art><m><sg>/$ ^phet<vblex><iv>/qaqee<vblex><iv>$<br />
* chunker: ^n<SN><aa><sg>{^masq<n><2><3>$}$ ^default<default>{^qaqee<vblex><iv>$}$<br />
* postchunk: ^masq<n><aa><sg>$ ^qaqee<vblex><iv>$<br />
* wam-kha: masq #qaqee (There is not a yet a structural transfer rule for subject agreement on verbs)<br />
<br />
masqak qaqeewak "The bears run"<br />
<br />
* tagger: ^masq<n><aa><pl>$ ^qaqee<vblex><iv><aa><p3><pl>$<br />
* biltrans: ^masq<n><aa><pl>/dngiem<n><m><pl>$ ^qaqee<vblex><iv><aa><p3><pl>/phet<vblex><iv><aa><p3><pl>$<br />
* chunker: ^art<SA><m><pl>{^article<art><3>$}$ ^n<SN><m><pl>{^dngiem<n><2>$}$ ^default<default>{^phet<vblex><iv><aa><p3><pl>$}$<br />
* postchunk: ^article<art><pl>$ ^dngiem<n><m>$ ^phet<vblex><iv><aa><p3><pl>$<br />
* wam-kha: ki dngiem #phet (Verb agreement rules are not implemented in this direction either)<br />
<br />
[[Category:Sp17_StructuralTransfer]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak/Structural_Transfer&diff=4365Khasi and Wôpanâak/Structural Transfer2017-04-07T03:25:37Z<p>Jmalin1: </p>
<hr />
<div>==Pre-Evaluation==<br />
===WER test===<br />
*13/17 correct<br />
===PER test===<br />
*23.53% unknown words<br />
===Coverage tests===<br />
Khasi<br />
*<br />
Wôpanâak<br />
*<br />
==Example Sentences==<br />
u dngiem u phet "The bear runs"<br />
<br />
tagger: ^article<art><m><sg>$ ^dngiem<n><m>$ ^article<art><m><sg>$ ^phet<vblex><iv>$<br />
biltrans: ^article<art><m><sg>/$ ^dngiem<n><m>/masq<n><aa>$ ^article<art><m><sg>/$ ^phet<vblex><iv>/qaqee<vblex><iv>$<br />
chunker: ^n<SN><aa><sg>{^masq<n><2><3>$}$ ^default<default>{^qaqee<vblex><iv>$}$<br />
postchunk: ^masq<n><aa><sg>$ ^qaqee<vblex><iv>$<br />
wam-kha: masq #qaqee (There is not a yet a structural transfer rule for subject agreement on verbs)<br />
<br />
masqak qaqeewak "The bears run"<br />
<br />
tagger: ^masq<n><aa><pl>$ ^qaqee<vblex><iv><aa><p3><pl>$<br />
biltrans: ^masq<n><aa><pl>/dngiem<n><m><pl>$ ^qaqee<vblex><iv><aa><p3><pl>/phet<vblex><iv><aa><p3><pl>$<br />
chunker: ^art<SA><m><pl>{^article<art><3>$}$ ^n<SN><m><pl>{^dngiem<n><2>$}$ ^default<default>{^phet<vblex><iv><aa><p3><pl>$}$<br />
postchunk: ^article<art><pl>$ ^dngiem<n><m>$ ^phet<vblex><iv><aa><p3><pl>$<br />
wam-kha: ki dngiem #phet (Verb agreement rules are not implemented in this direction either)<br />
<br />
[[Category:Sp17_StructuralTransfer]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak/Lexical_Selection&diff=4089Khasi and Wôpanâak/Lexical Selection2017-03-28T16:35:53Z<p>Jmalin1: </p>
<hr />
<div>===wam-->kha===<br />
*meequn "feather, pen" => sner "feather", khulom "pen", khulom-sner "feather pen"<br />
*keesuk "sky, a day" => sngi "day, sun", bneng "sky"<br />
*nupâus "sun, month" => sngi "day, sun", <br />
<br />
===kha-->wam===<br />
* kjat "foot/leg" => -ukhôt "leg", -seet "foot"<br />
* ngi "we (1st person plural pronoun)" => neenawun "we (exclusive)", keenawun "we (inclusive)" This is reflective of a deeper grammatical ambiguity with regards to first person plural forms throughout the languages.<br />
* sngi "day, sun" => keesuk "sky, day", nupâus "sun, month"<br />
<br />
[[Category:Sp17_LexicalSelection]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak/Lexical_Selection&diff=4065Khasi and Wôpanâak/Lexical Selection2017-03-28T16:08:09Z<p>Jmalin1: </p>
<hr />
<div>===wam-->kha===<br />
*meequn "feather, pen" => sner "feather", khulom "pen", khulom-sner "feather pen"<br />
*keesuk "sky, a day" => sngi "day, sun", bneng "sky"<br />
<br />
===kha-->wam===<br />
* kjat "foot/leg" => -ukhôt "leg", -seet "foot"<br />
* ngi "we (1st person plural pronoun)" => neenawun "we (exclusive)", keenawun "we (inclusive)" This is reflective of a deeper grammatical ambiguity with regards to first person plural forms throughout the languages.<br />
<br />
[[Category:Sp17_LexicalSelection]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak&diff=4026Khasi and Wôpanâak2017-03-28T03:22:45Z<p>Jmalin1: /* wam-kha Evaluation */</p>
<hr />
<div>Resources for machine translation between [[Khasi|Khasi]] and [[Wôpanâak|Wôpanâak]]<br />
* [[Khasi_and_Wôpanâak/ContrastiveGrammars|Contrastive Grammars]]<br />
<br />
[[Category:Sp17_TranslationPairs]]<br />
<br />
==Lexical Selection==<br />
Here are documented one-to-many mappings in either direction for the language pair.<br />
===Khasi to Wôpanâak===<br />
kjat "foot/leg" => -ukhôt "leg", -seet "foot"<br />
<br />
ngi "we (1st person plural pronoun)" => neenawun "we (exclusive)", keenawun "we (inclusive)" This is reflective of a deeper grammatical ambiguity with regards to first person plural forms throughout the languages.<br />
<br />
===Wôpanâak to Khasi===<br />
meequn "feather, pen" => sner "feather", khulom "pen", khulom-sner "feather pen"<br />
<br />
keesuk "sky, a day" => sngi "day, sun", bneng "sky"<br />
<br />
==wam-kha Evaluation==<br />
Current WER is 93.75%. Current PER is 93.75%.<br />
12 out of 15 stems are translating correctly. The remaining stems are mô and weeqayach. The first is not translating because the Wôpanâak copula does not currently have a root lemma, and the second is due to a difference in the way the two texts realize "Let there be light" - in Wôpanâak this is expressed with a single verb, while in Khasi it is closer to "let light be".</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak/ContrastiveGrammars&diff=4025Khasi and Wôpanâak/ContrastiveGrammars2017-03-28T03:12:38Z<p>Jmalin1: </p>
<hr />
<div>=Differences in Grammar (with tests)=<br />
Wôpanâak has an animate/inanimate distinction for verbs, while Khasi does not<br />
* {{transferTest|kha|wam|thaw|ayum}}<br />
* {{transferMorphTest|kha|wam|thaw{{tag|v}}|ayum{{tag|v}}{{tag|aa}}{{tag|tv}}{{tag|p3}}{{tag|sg}}}}<br />
<br />
* {{transferTest|wam|kha|ayum|thaw}}<br />
* {{transferMorphTest|wam|kha|ayum{{tag|v}}{{tag|aa}}{{tag|tv}}{{tag|p3}}{{tag|sg}}|thaw{{tag|v}}}}<br />
Wôpanâak does not have articles, while Khasi does<br />
* {{transferTest|kha|wam|ka bneng|keesuk}}<br />
* {{transferMorphTest|kha|wam|ka{{tag|art}}{{tag|f}}{{tag|sg}} bneng{{tag|n}}|keesuk{{tag|n}}{{tag|nn}}{{tag|sg}}}}<br />
<br />
* {{transferTest|wam|kha|keesuk|ka bneng}}<br />
* {{transferMorphTest|wam|kha|keesuk{{tag|n}}{{tag|nn}}{{tag|sg}}|ka{{tag|art}}{{tag|f}}{{tag|sg}} bneng{{tag|n}}}}<br />
Wôpanâak has morphology determining person and number for verbs, while Khasi does not<br />
* {{transferTest|kha|wam|thaw|ayum}}<br />
* {{transferMorphTest|kha|wam|thaw{{tag|v}}|ayum{{tag|v}}{{tag|aa}}{{tag|tv}}{{tag|p3}}{{tag|sg}}}}<br />
<br />
* {{transferTest|wam|kha|ayum|thaw}}<br />
* {{transferMorphTest|wam|kha|ayum{{tag|v}}{{tag|aa}}{{tag|tv}}{{tag|p3}}{{tag|sg}}|thaw{{tag|v}}}}<br />
Wôpanâak has morphology determining tense, while Khasi has a separate particle<br />
* {{transferTest|kha|wam|la long|mô}}<br />
* {{transferMorphTest|kha|wam|la{{tag|pastPar}} long{{tag|v}}|mô{{tag|v}}{{tag|indef}}{{tag|past}}}}<br />
<br />
* {{transferTest|wam|kha|mô|la long}}<br />
* {{transferMorphTest|wam|kha|mô{{tag|v}}{{tag|indef}}{{tag|past}}|la{{tag|pastPar}} long{{tag|v}}}}<br />
Khasi is SVO, while word order in Wôpanâak is somewhat arbitrary (and somewhat VSO)<br />
* {{transferTest|kha|wam|U Blei u la ong, to long ka jingshai, bad ka jingshai ka la long|nȣwâw God weeqayach kah mô weeqay}}<br />
* {{transferMorphTest|kha|wam|U{{tag|art}}{{tag|m}}{{tag|sg}} Blei{{tag|n}} u{{tag|art}}{{tag|m}}{{tag|sg}} la{{tag|pastPar}} ong{{tag|v}} to{{tag|imp}}{{tag|v}} long{{tag|v}} ka{{tag|art}}{{tag|f}}{{tag|sg}} jingshai{{tag|n}} bad{{tag|coocnj}} ka{{tag|art}}{{tag|f}}{{tag|sg}} jingshai{{tag|n}} ka{{tag|art}}{{tag|f}}{{tag|sg}} la{{tag|pastPar}} long{{tag|v}}|nȣwâw {{tag|v}}{{tag|aa}}{{tag|iv}}{{tag|p3}}{{tag|sg}} God{{tag|n}}{{tag|aa}}{{tag|sg}} weeqayach{{tag|v}}{{tag|nn}}{{tag|imp}}{{tag|p3}}{{tag|sg}} kah{{tag|coocnj}} mô{{tag|v}}{{tag|indef}}{{tag|past}} weeqay{{tag|n}}{{tag|nn}}{{tag|sg}}}}<br />
<br />
*{{transferTest|wam|kha|nȣwâw God weeqayach kah mô weeqay|U Blei u la ong, to long ka jingshai, bad ka jingshai ka la long}}<br />
*{{transferMorphTest|wam|kha|nȣwâw{{tag|v}}{{tag|aa}}{{tag|iv}}{{tag|p3}}{{tag|sg}} God{{tag|n}}{{tag|aa}}{{tag|sg}} weeqayach{{tag|v}}{{tag|nn}}{{tag|imp}}{{tag|p3}}{{tag|sg}} kah{{tag|coocnj}} mô{{tag|v}}{{tag|indef}}{{tag|past}} weeqay{{tag|n}}{{tag|nn}}{{tag|sg}}| U{{tag|art}}{{tag|m}}{{tag|sg}} Blei{{tag|n}} u{{tag|art}}{{tag|m}}{{tag|sg}} la{{tag|pastPar}} ong{{tag|v}} to{{tag|imp}}{{tag|v}} long{{tag|v}} ka{{tag|art}}{{tag|f}}{{tag|sg}} jingshai{{tag|n}} bad{{tag|coocnj}} ka{{tag|art}}{{tag|f}}{{tag|sg}} jingshai{{tag|n}} ka{{tag|art}}{{tag|f}}{{tag|sg}} la{{tag|pastPar}} long{{tag|v}}}}<br />
<br />
== Further Kha-Wam test ==<br />
{{transferTest|kha|wam|U Blei la thaw ia ka bneng bad ia ka khyndew|ayum God keesuk kah ahkee}}<br />
<br />
{{transferMorphTest|kha|wam|U{{tag|art}}{{tag|m}}{{tag|sg}} Blei{{tag|n}} la{{tag|pastPar}} thaw{{tag|v}} ia{{tag|dirObj}} ka{{tag|art}}{{tag|f}}{{tag|sg}} bneng{{tag|n}} bad{{tag|coocnj}} ia{{tag|dirObj}} ka{{tag|art}}{{tag|f}}{{tag|sg}} khyndew{{tag|n}}|ayum{{tag|v}}{{tag|aa}}{{tag|tv}}{{tag|p3}}{{tag|sg}} God{{tag|n}}{{tag|aa}}{{tag|sg}} keesuk{{tag|n}}{{tag|nn}}{{tag|sg}} kah{{tag|coocnj}} ahkee{{tag|n}}{{tag|n}}{{tag|sg}}}}<br />
<br />
*English gloss: God created heaven and earth<br />
<br />
== Further Wam-Kha test ==<br />
{{transferTest|wam|kha|ayum God keesuk kah ahkee|U Blei la thaw ia ka bneng bad ia ka khyndew}}<br />
<br />
{{transferMorphTest|wam|kha|ayum{{tag|v}}{{tag|aa}}{{tag|tv}}{{tag|p3}}{{tag|sg}} God{{tag|n}}{{tag|aa}}{{tag|sg}} keesuk{{tag|n}}{{tag|nn}}{{tag|sg}} kah{{tag|coocnj}} ahkee{{tag|n}}{{tag|n}}{{tag|sg}}|U{{tag|art}}{{tag|m}}{{tag|sg}} Blei{{tag|n}} la{{tag|pastPar}} thaw{{tag|v}} ia{{tag|dirObj}} ka{{tag|art}}{{tag|f}}{{tag|sg}} bneng{{tag|n}} bad{{tag|coocnj}} ia{{tag|dirObj}} ka{{tag|art}}{{tag|f}}{{tag|sg}} khyndew{{tag|n}}}}<br />
<br />
*English gloss: God created heaven and earth<br />
<br />
[[Category:Sp17_ContrastiveGrammars]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak&diff=4024Khasi and Wôpanâak2017-03-28T03:08:55Z<p>Jmalin1: /* wam-kha Evaluation */</p>
<hr />
<div>Resources for machine translation between [[Khasi|Khasi]] and [[Wôpanâak|Wôpanâak]]<br />
* [[Khasi_and_Wôpanâak/ContrastiveGrammars|Contrastive Grammars]]<br />
<br />
[[Category:Sp17_TranslationPairs]]<br />
<br />
==Lexical Selection==<br />
Here are documented one-to-many mappings in either direction for the language pair.<br />
===Khasi to Wôpanâak===<br />
kjat "foot/leg" => -ukhôt "leg", -seet "foot"<br />
<br />
ngi "we (1st person plural pronoun)" => neenawun "we (exclusive)", keenawun "we (inclusive)" This is reflective of a deeper grammatical ambiguity with regards to first person plural forms throughout the languages.<br />
<br />
===Wôpanâak to Khasi===<br />
meequn "feather, pen" => sner "feather", khulom "pen", khulom-sner "feather pen"<br />
<br />
keesuk "sky, a day" => sngi "day, sun", bneng "sky"<br />
<br />
==wam-kha Evaluation==<br />
Current WER is 93.75%. Current PER is 93.75%.<br />
10 out of 16 stems are translating correctly</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak&diff=4023Khasi and Wôpanâak2017-03-28T02:54:12Z<p>Jmalin1: </p>
<hr />
<div>Resources for machine translation between [[Khasi|Khasi]] and [[Wôpanâak|Wôpanâak]]<br />
* [[Khasi_and_Wôpanâak/ContrastiveGrammars|Contrastive Grammars]]<br />
<br />
[[Category:Sp17_TranslationPairs]]<br />
<br />
==Lexical Selection==<br />
Here are documented one-to-many mappings in either direction for the language pair.<br />
===Khasi to Wôpanâak===<br />
kjat "foot/leg" => -ukhôt "leg", -seet "foot"<br />
<br />
ngi "we (1st person plural pronoun)" => neenawun "we (exclusive)", keenawun "we (inclusive)" This is reflective of a deeper grammatical ambiguity with regards to first person plural forms throughout the languages.<br />
<br />
===Wôpanâak to Khasi===<br />
meequn "feather, pen" => sner "feather", khulom "pen", khulom-sner "feather pen"<br />
<br />
keesuk "sky, a day" => sngi "day, sun", bneng "sky"<br />
<br />
==wam-kha Evaluation==<br />
Current WER is 93.75%. Current PER is 93.75%.</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak&diff=4021Khasi and Wôpanâak2017-03-28T02:41:53Z<p>Jmalin1: </p>
<hr />
<div>Resources for machine translation between [[Khasi|Khasi]] and [[Wôpanâak|Wôpanâak]]<br />
* [[Khasi_and_Wôpanâak/ContrastiveGrammars|Contrastive Grammars]]<br />
<br />
[[Category:Sp17_TranslationPairs]]<br />
<br />
==Lexical Selection==<br />
Here are documented one-to-many mappings in either direction for the language pair.<br />
===Khasi to Wôpanâak===<br />
kjat "foot/leg" => -ukhôt "leg", -seet "foot"<br />
<br />
ngi "we (1st person plural pronoun)" => neenawun "we (exclusive)", keenawun "we (inclusive)" This is reflective of a deeper grammatical ambiguity with regards to first person plural forms throughout the languages.<br />
<br />
===Wôpanâak to Khasi===<br />
meequn "feather, pen" => sner "feather", khulom "pen", khulom-sner "feather pen"<br />
<br />
keesuk "sky, a day" => sngi "day, sun", bneng "sky"</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=Khasi_and_W%C3%B4pan%C3%A2ak&diff=4020Khasi and Wôpanâak2017-03-28T02:41:33Z<p>Jmalin1: </p>
<hr />
<div>Resources for machine translation between [[Khasi|Khasi]] and [[Wôpanâak|Wôpanâak]]<br />
* [[Khasi_and_Wôpanâak/ContrastiveGrammars|Contrastive Grammars]]<br />
<br />
[[Category:Sp17_TranslationPairs]]<br />
<br />
==Lexical Selection==<br />
Here are documented one-to-many mappings in either direction for the language pair.<br />
===Khasi to Wôpanâak===<br />
kjat "foot/leg" => -ukhôt "leg", -seet "foot"<br />
ngi "we (1st person plural pronoun)" => neenawun "we (exclusive)", keenawun "we (inclusive)" This is reflective of a deeper grammatical ambiguity with regards to first person plural forms throughout the languages.<br />
<br />
===Wôpanâak to Khasi===<br />
meequn "feather, pen" => sner "feather", khulom "pen", khulom-sner "feather pen"<br />
keesuk "sky, a day" => sngi "day, sun", bneng "sky"</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak/Disambiguation&diff=3312Wôpanâak/Disambiguation2017-03-02T05:07:00Z<p>Jmalin1: </p>
<hr />
<div>==Initial Evaluation of Ambiguity==<br />
Initial ambiguity is calculated at 1.026. However, this analysis may be impacted by tokenisation errors.<br />
<br />
[[Category:Wôpanâak]][[Category:Sp17_Disambiguation]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak/Disambiguation&diff=3306Wôpanâak/Disambiguation2017-03-02T04:48:54Z<p>Jmalin1: </p>
<hr />
<div>==Initial Evaluation of Ambiguity==<br />
Initial ambiguity is calculated at 1.039. However, this analysis may be impacted by tokenisation errors.<br />
<br />
[[Category:Wôpanâak]][[Category:Sp17_Disambiguation]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak/Disambiguation&diff=3305Wôpanâak/Disambiguation2017-03-02T04:15:53Z<p>Jmalin1: Created page with "==Initial Evaluation of Ambiguity== Category:Sp17_Disambiguation"</p>
<hr />
<div>==Initial Evaluation of Ambiguity==<br />
<br />
[[Category:Sp17_Disambiguation]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak&diff=3304Wôpanâak2017-03-02T04:14:34Z<p>Jmalin1: </p>
<hr />
<div>==External Resources==<br />
<br />
===Grammar===<br />
<br />
A grammar of the language published in 2000 by Jessie Little Doe Fermino (or Jessie Little Doe Baird) as an MIT thesis is available online [http://dspace.mit.edu/handle/1721.1/8740 here]. Most other resources I have found at this point are the sources used to develop this grammar. It is under copyright.<br />
<br />
Another [http://dspace.mit.edu/handle/1721.1/37422 thesis], by Nitana Hicks, lists morphemes used in Wôpanâak verbs along with rules for verb construction.<br />
<br />
Additionally, Volume 2 of ''Native Writings in Massachusett'' contains a "Grammatical Sketch" based on several centuries-older grammars.<br />
<br />
===Dictionary===<br />
<br />
The Natick Dictionary, published in 1903, appears to be a dictionary of at least one form of the language available online [https://archive.org/details/natickdictionar02trumgoog here]. It is a dictionary between English and the Natick dialect of Wôpanâak.<br />
<br />
===Corpora===<br />
<br />
The 1988 book ''Native Writings In Massachusett'' (available from McCabe and online), by Ives Goddard and Kathleen Bragdon, is the primary source of text for the language, containing many (possibly all) surviving texts written by native speakers of the language from the 1660s to the 1750s. The book contains many documents with transcriptions, annotations, and translations. The documents are in large part legal documents, including land transactions and wills, although documents with other purposes are in the collection as well. The book was published by the American Philosophical Society and is copyrighted.<br />
<br />
Additionally, a major source for the language is a translation of the Bible written by a non-native speaker (missionary John Eliot). Editions from [https://archive.org/details/mamussewunneetup00elio 1663] and [https://archive.org/details/mamussewunneetup02elio 1685] are available online, although they appear to be in image format without good transcription.<br />
<br />
==Developed Resources==<br />
<br />
===Keyboard Layouts===<br />
<br />
Preliminary [[Wôpanâak/Keyboard|keyboard layouts]] for Wôpanâak are available on [https://github.swarthmore.edu/jmalin1/ling073-wam-keyboard GitHub].<br />
<br />
[[Category:Wôpanâak]] [[Category:Sp17 ResourceDocumentation]]<br />
<br />
===Corpora===<br />
<br />
Basic corpora are available [https://github.swarthmore.edu/jmalin1/ling073-wam-corpus/graphs/contributors here] (still being constructed). The repository is private due to copyrighted sources for the corpora.<br />
<br />
===Morphological Analyzer===<br />
<br />
Some morphological processes are documented [[Wôpanâak/Grammar|here]].<br />
<br />
A preliminary morphological analyzer is under development [https://github.swarthmore.edu/jmalin1/ling073-wam here]. Currently it mainly analyzes nouns. More information can be found at [[Wôpanâak/Transducer]].<br />
<br />
===Disambiguation===<br />
<br />
A disambiguator is under development at [[Wôpanâak/Disambiguation]].</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak&diff=3303Wôpanâak2017-03-02T04:14:25Z<p>Jmalin1: </p>
<hr />
<div>==External Resources==<br />
<br />
===Grammar===<br />
<br />
A grammar of the language published in 2000 by Jessie Little Doe Fermino (or Jessie Little Doe Baird) as an MIT thesis is available online [http://dspace.mit.edu/handle/1721.1/8740 here]. Most other resources I have found at this point are the sources used to develop this grammar. It is under copyright.<br />
<br />
Another [http://dspace.mit.edu/handle/1721.1/37422 thesis], by Nitana Hicks, lists morphemes used in Wôpanâak verbs along with rules for verb construction.<br />
<br />
Additionally, Volume 2 of ''Native Writings in Massachusett'' contains a "Grammatical Sketch" based on several centuries-older grammars.<br />
<br />
===Dictionary===<br />
<br />
The Natick Dictionary, published in 1903, appears to be a dictionary of at least one form of the language available online [https://archive.org/details/natickdictionar02trumgoog here]. It is a dictionary between English and the Natick dialect of Wôpanâak.<br />
<br />
===Corpora===<br />
<br />
The 1988 book ''Native Writings In Massachusett'' (available from McCabe and online), by Ives Goddard and Kathleen Bragdon, is the primary source of text for the language, containing many (possibly all) surviving texts written by native speakers of the language from the 1660s to the 1750s. The book contains many documents with transcriptions, annotations, and translations. The documents are in large part legal documents, including land transactions and wills, although documents with other purposes are in the collection as well. The book was published by the American Philosophical Society and is copyrighted.<br />
<br />
Additionally, a major source for the language is a translation of the Bible written by a non-native speaker (missionary John Eliot). Editions from [https://archive.org/details/mamussewunneetup00elio 1663] and [https://archive.org/details/mamussewunneetup02elio 1685] are available online, although they appear to be in image format without good transcription.<br />
<br />
==Developed Resources==<br />
<br />
===Keyboard Layouts===<br />
<br />
Preliminary [[Wôpanâak/Keyboard|keyboard layouts]] for Wôpanâak are available on [https://github.swarthmore.edu/jmalin1/ling073-wam-keyboard GitHub].<br />
<br />
[[Category:Wôpanâak]] [[Category:Sp17 ResourceDocumentation]]<br />
<br />
===Corpora===<br />
<br />
Basic corpora are available [https://github.swarthmore.edu/jmalin1/ling073-wam-corpus/graphs/contributors here] (still being constructed). The repository is private due to copyrighted sources for the corpora.<br />
<br />
===Morphological Analyzer===<br />
<br />
Some morphological processes are documented [[Wôpanâak/Grammar|here]].<br />
<br />
A preliminary morphological analyzer is under development [https://github.swarthmore.edu/jmalin1/ling073-wam here]. Currently it mainly analyzes nouns. More information can be found at [[Wôpanâak/Transducer]].<br />
<br />
===Disambiguation===<br />
<br />
A disambiguator is under development at [[Wôpanâak/Disambiguation]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak/Transducer&diff=3084Wôpanâak/Transducer2017-02-23T05:50:55Z<p>Jmalin1: /* Current Evaluation */</p>
<hr />
<div>An in-progress transducer is available [https://github.swarthmore.edu/jmalin1/ling073-wam here]<br />
<br />
==Evaluation==<br />
<br />
===Analyzer Evaluation===<br />
The transducer contained 23 different correctly inflecting noun roots, and one incompletely analyzed verb root. The current corpus coverage is 15.44% (but see below).<br />
<br />
The top corpus words not yet entered were weeyâws, nupȣnum, wutche, puhpeeq, nuhtatôm, nâwâw, nunâw, môcheenat, and nuwachônuqun. (Some short words are excluded from this list because they are likely functional words taken from older texts with nonstandard orthography.)<br />
<br />
In the current yaml file, 55 tests pass and 22 fail.<br />
<br />
===Generator Evaluation===<br />
====Previous Analyser Evaluation====<br />
At the start of this assignment, 55 out of 88 tests passed. Some of the failures were due to nouns that were analyzing correctly but generating wrong, while many were due to verbs not being implemented yet.<br />
<br />
====Current Evaluation====<br />
In the current yaml file, 68 tests pass the generation test and 9 fail. The failing tests are all pronouns and other things that have not been implemented - the tests for nouns and intransitive verbs are all working. However, the transducer cannot handle many things not present in the tests.<br />
<br />
I added 9 twol rules. 6 of these were rules for noun generation, which was implemented before this assignment, and 3 are for verbs, which I only began to implement this week.<br />
<br />
Corpus coverage is currently 20.1%.<br />
<br />
==Notes==<br />
This analyzer currently analyzes nouns. As of yet, it can properly analyze nouns for possession, number, animacy, and whether the noun is absentative, obviative, or locative. (All these analyses are based on Baird's grammar.) Not too many nouns have been entered into the lexicon yet, and in particular few nouns with truncating stems. At the moment, some of these overgenerate due to the presence of an archiphoneme, while others are sent through special lexicons to get the material before any suffixes.<br />
<br />
===Corpus Coverage===<br />
The corpus has many issues, the largest of which being that for much of the text orthography is not standardized. The analysis successes all come from sentences taken from the grammar.<br />
<br />
Initial corpus coverage was 10.59%. Adding "waskeetôp<n><aa><sg>" and "seepȣ<n><nn><sg>", and fixing "mahkus" so that its stem inflected as reflected in the corpus, improved coverage to 15.44%.<br />
<br />
[[Category:Wôpanâak]][[Category:Sp17_Transducers]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak/Transducer&diff=3082Wôpanâak/Transducer2017-02-23T05:35:33Z<p>Jmalin1: </p>
<hr />
<div>An in-progress transducer is available [https://github.swarthmore.edu/jmalin1/ling073-wam here]<br />
<br />
==Evaluation==<br />
<br />
===Analyzer Evaluation===<br />
The transducer contained 23 different correctly inflecting noun roots, and one incompletely analyzed verb root. The current corpus coverage is 15.44% (but see below).<br />
<br />
The top corpus words not yet entered were weeyâws, nupȣnum, wutche, puhpeeq, nuhtatôm, nâwâw, nunâw, môcheenat, and nuwachônuqun. (Some short words are excluded from this list because they are likely functional words taken from older texts with nonstandard orthography.)<br />
<br />
In the current yaml file, 55 tests pass and 22 fail.<br />
<br />
===Generator Evaluation===<br />
====Previous Analyser Evaluation====<br />
At the start of this assignment, 55 out of 88 tests passed. Some of the failures were due to nouns that were analyzing correctly but generating wrong, while many were due to verbs not being implemented yet.<br />
<br />
====Current Evaluation====<br />
In the current yaml file, 68 tests pass the generation test and 9 fail. The failing tests are all pronouns and other things that have not been implemented - the tests for nouns and intransitive verbs are all working. However, the transducer cannot handle many things not present in the tests.<br />
<br />
I added 9 twol rules. 6 of these were rules for noun generation, which was implemented before this assignment, and 3 are for verbs, which I only began to implement this week.<br />
<br />
Corpus coverage is not currently working, for some reason.<br />
<br />
==Notes==<br />
This analyzer currently analyzes nouns. As of yet, it can properly analyze nouns for possession, number, animacy, and whether the noun is absentative, obviative, or locative. (All these analyses are based on Baird's grammar.) Not too many nouns have been entered into the lexicon yet, and in particular few nouns with truncating stems. At the moment, some of these overgenerate due to the presence of an archiphoneme, while others are sent through special lexicons to get the material before any suffixes.<br />
<br />
===Corpus Coverage===<br />
The corpus has many issues, the largest of which being that for much of the text orthography is not standardized. The analysis successes all come from sentences taken from the grammar.<br />
<br />
Initial corpus coverage was 10.59%. Adding "waskeetôp<n><aa><sg>" and "seepȣ<n><nn><sg>", and fixing "mahkus" so that its stem inflected as reflected in the corpus, improved coverage to 15.44%.<br />
<br />
[[Category:Wôpanâak]][[Category:Sp17_Transducers]]</div>Jmalin1http://wikis.swarthmore.edu/ling073/index.php?title=W%C3%B4pan%C3%A2ak/Transducer&diff=3078Wôpanâak/Transducer2017-02-23T04:57:51Z<p>Jmalin1: </p>
<hr />
<div>An in-progress transducer is available [https://github.swarthmore.edu/jmalin1/ling073-wam here]<br />
<br />
==Evaluation==<br />
<br />
===Analyzer Evaluation===<br />
The transducer contained 23 different correctly inflecting noun roots, and one incompletely analyzed verb root. The current corpus coverage is 15.44% (but see below).<br />
<br />
The top corpus words not yet entered were weeyâws, nupȣnum, wutche, puhpeeq, nuhtatôm, nâwâw, nunâw, môcheenat, and nuwachônuqun. (Some short words are excluded from this list because they are likely functional words taken from older texts with nonstandard orthography.)<br />
<br />
In the current yaml file, 55 tests pass and 22 fail.<br />
<br />
===Generator Evaluation===<br />
In the current yaml file, 68 tests pass the generation test and 9 fail.<br />
<br />
I added 9 twol rules. 6 of these were rules for noun generation, which was implemented before this assignment, and 3 are for verbs, which I only began to implement this week.<br />
<br />
Corpus coverage is not currently working, for some reason.<br />
<br />
==Notes==<br />
This analyzer currently analyzes nouns. As of yet, it can properly analyze nouns for possession, number, animacy, and whether the noun is absentative, obviative, or locative. (All these analyses are based on Baird's grammar.) Not too many nouns have been entered into the lexicon yet, and in particular few nouns with truncating stems. At the moment, some of these overgenerate due to the presence of an archiphoneme, while others are sent through special lexicons to get the material before any suffixes.<br />
<br />
===Corpus Coverage===<br />
The corpus has many issues, the largest of which being that for much of the text orthography is not standardized. The analysis successes all come from sentences taken from the grammar.<br />
<br />
Initial corpus coverage was 10.59%. Adding "waskeetôp<n><aa><sg>" and "seepȣ<n><nn><sg>", and fixing "mahkus" so that its stem inflected as reflected in the corpus, improved coverage to 15.44%.<br />
<br />
[[Category:Wôpanâak]][[Category:Sp17_Transducers]]</div>Jmalin1