Difference between revisions of "Maasai/Final Project"

From LING073
Jump to: navigation, search
m (→‎Present Tense)
 
(47 intermediate revisions by 3 users not shown)
Line 1: Line 1:
 
=Transducer for Maasai=
 
=Transducer for Maasai=
 +
Transducer Repo: <code>https://github.com/kdandy1/ling073-mas </code>
  
−
Code <code>https://github.swarthmore.edu/Ling073-sp23/ling073-mas</code>
+
==Classes/Tenses Added==
 +
*Nouns: nominative, accusative, vocative
 +
*Adjectives
 +
*Verbs: present, imperfect, past
  
−
==Classes/Tenses Added==
+
==Morphtests==
−
*  
+
 
 +
===Nouns===
 +
Note: all of the grammar books just say that there is no set rule to form the plural of nouns (or adjectives), so just memorize, so we're gonna do that too.
 +
 
 +
====Nominative====
 +
 
 +
Feminine: en prefix in the singular and in in the plural
 +
*{{morphTest|tito{{tag|n}}{{tag|f}}{{tag|sg}}{{tag|nom}}|entito}}
 +
*{{morphTest|tito{{tag|n}}{{tag|f}}{{tag|pl}}{{tag|nom}}|intoyie}}
 +
*{{morphTest|kitok{{tag|n}}{{tag|f}}{{tag|sg}}{{tag|nom}}|enkitok}}
 +
*{{morphTest|kitok{{tag|n}}{{tag|f}}{{tag|pl}}{{tag|nom}}|inkituaak}}
 +
*{{morphTest|jore{{tag|n}}{{tag|f}}{{tag|sg}}{{tag|nom}}|enjore}}
 +
*{{morphTest|jore{{tag|n}}{{tag|f}}{{tag|pl}}{{tag|nom}}|injorin}}
 +
*{{morphTest|kima{{tag|n}}{{tag|f}}{{tag|sg}}{{tag|nom}}|enkima}} 
 +
*{{morphTest|kima{{tag|n}}{{tag|f}}{{tag|pl}}{{tag|nom}}|inkimaitie}}
 +
*{{morphTest|aina{{tag|n}}{{tag|f}}{{tag|sg}}{{tag|nom}}|enkaina}}
 +
*{{morphTest|aina{{tag|n}}{{tag|f}}{{tag|pl}}{{tag|nom}}|inkaik}}
 +
 
 +
Masculine: ol prefix in the singular and il in the plural
 +
*{{morphTest|ayioni{{tag|n}}{{tag|m}}{{tag|sg}}{{tag|nom}}|olayioni}}
 +
*{{morphTest|ayioni{{tag|n}}{{tag|m}}{{tag|pl}}{{tag|nom}}|ilayiok}}
 +
*{{morphTest|payian{{tag|n}}{{tag|m}}{{tag|sg}}{{tag|nom}}|olpayian}}
 +
*{{morphTest|payian{{tag|n}}{{tag|m}}{{tag|pl}}{{tag|nom}}|ilpayiani}}
 +
*{{morphTest|lee{{tag|n}}{{tag|m}}{{tag|sg}}{{tag|nom}}|olee}}
 +
*{{morphTest|lee{{tag|n}}{{tag|m}}{{tag|pl}}{{tag|nom}}|ilewa}}
 +
*{{morphTest|soit{{tag|n}}{{tag|m}}{{tag|sg}}{{tag|nom}}|osoit}} 
 +
*{{morphTest|soit{{tag|n}}{{tag|m}}{{tag|pl}}{{tag|nom}}|isoito}}
 +
*{{morphTest|reyiet{{tag|n}}{{tag|m}}{{tag|sg}}{{tag|nom}}|oreyiet}}
 +
*{{morphTest|reyiet{{tag|n}}{{tag|m}}{{tag|pl}}{{tag|nom}}|ireyieta}}
 +
 
 +
Neuter: only one noun is neuter haha, prefixes are e in the singular and i in the plural
 +
*{{morphTest|wueji{{tag|n}}{{tag|nt}}{{tag|pl}}{{tag|nom}}|ewueji}}
 +
*{{morphTest|wueji{{tag|n}}{{tag|nt}}{{tag|pl}}{{tag|nom}}|iwuejitin}}
  
−
==Analyser Evaluation==
+
===Adjectives===
 +
*{{morphTest|arriyia{{tag|adj}}{{tag|sg}}|arriyia}}   
 +
*{{morphTest|arriyia{{tag|adj}}{{tag|pl}}|arriyak}}
 +
*{{morphTest|bioto{{tag|adj}}{{tag|sg}}|bioto}}
 +
*{{morphTest|bioto{{tag|adj}}{{tag|pl}}|biot}}
 +
*{{morphTest|dalut{{tag|adj}}{{tag|sg}}|dalut}}
 +
*{{morphTest|dalut{{tag|adj}}{{tag|pl}}|daluti}}
 +
*{{morphTest|asipani{{tag|adj}}{{tag|sg}}|asipani}}
 +
*{{morphTest|asipani{{tag|adj}}{{tag|pl}}|asipak}}
 +
*{{morphTest|kitok{{tag|adj}}{{tag|sg}}|kitok}}
 +
*{{morphTest|kitok{{tag|adj}}{{tag|pl}}|kituaak}}
  
−
===Precision and Recall===
+
===Verbs===
−
Totals:
 
−
Precision:
 
−
Recall:
 
  
−
===Current Coverage===
+
====Present Tense====
−
coverage:
+
prefix + root
 +
"to follow"
 +
*{{morphTest|suj{{tag|vblex}}{{tag|prs}}{{tag|p1}}{{tag|sg}}|asuj}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|prs}}{{tag|p1}}{{tag|pl}}|kisuj}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|prs}}{{tag|p2}}{{tag|sg}}|isuj}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|prs}}{{tag|p2}}{{tag|pl}}|isujusuju}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|prs}}{{tag|p3}}{{tag|sg}}|esuj}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|prs}}{{tag|p3}}{{tag|pl}}|esuj}}
 +
"to wash"
 +
*{{morphTest|isuj{{tag|vblex}}{{tag|prs}}{{tag|p1}}{{tag|sg}}|aisuj}}
 +
*{{morphTest|isuj{{tag|vblex}}{{tag|prs}}{{tag|p1}}{{tag|pl}}|kiisuj}}
 +
*{{morphTest|isuj{{tag|vblex}}{{tag|prs}}{{tag|p2}}{{tag|sg}}|iisuj}}
 +
*{{morphTest|isuj{{tag|vblex}}{{tag|prs}}{{tag|p2}}{{tag|pl}}|iisujuisuju}}
 +
*{{morphTest|isuj{{tag|vblex}}{{tag|prs}}{{tag|p3}}{{tag|sg}}|eisuj}}
 +
*{{morphTest|isuj{{tag|vblex}}{{tag|prs}}{{tag|p3}}{{tag|pl}}|eisuj}}
  
−
remaining unknown forms:
+
====Imperfect Progressive====
 +
prefix + root + suffix
 +
"to follow"
 +
*{{morphTest|suj{{tag|vblex}}{{tag|prs_impf}}{{tag|p1}}{{tag|sg}}|asujita}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|prs_impf}}{{tag|p1}}{{tag|pl}}|kisujita}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|prs_impf}}{{tag|p2}}{{tag|sg}}|isujita}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|prs_impf}}{{tag|p2}}{{tag|pl}}|isujita}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|prs_impf}}{{tag|p3}}{{tag|sg}}|esujita}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|prs_impf}}{{tag|p3}}{{tag|pl}}|esujita}}
  
−
ambiguity:
+
====Past====
 +
a + (t vowel) + root + suffix
  
−
===Number of Stems in the Transducer===
+
various changes based on root vowel being open and closed
−
Lexicons:
+
*{{morphTest|suj{{tag|vblex}}{{tag|past}}{{tag|p1}}{{tag|sg}}|atusuja}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|past}}{{tag|p1}}{{tag|pl}}|kitusuja}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|past}}{{tag|p2}}{{tag|sg}}|itusuja}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|past}}{{tag|p2}}{{tag|pl}}|itusuja}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|past}}{{tag|p3}}{{tag|sg}}|etusuja}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|past}}{{tag|p3}}{{tag|pl}}|etusuja}}
  
−
Lexicon entries:
+
====Narrative====
 +
n + vowel + present tense form
 +
*{{morphTest|suj{{tag|vblex}}{{tag|narr}}{{tag|p1}}{{tag|sg}}|nasuj}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|narr}}{{tag|p1}}{{tag|pl}}|nikisuj}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|narr}}{{tag|p2}}{{tag|sg}}|nisuj}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|narr}}{{tag|p2}}{{tag|pl}}|nisujusuju}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|narr}}{{tag|p3}}{{tag|sg}}|nesuj}}
 +
*{{morphTest|suj{{tag|vblex}}{{tag|narr}}{{tag|p3}}{{tag|pl}}|nesuj}}
  
−
Patterns:
+
==Analyser Evaluation==
  
−
Pattern entries:  
+
===Precision and Recall===
 +
Totals: 152 forms, 199 tp, 8 fp, 0 tn, 111 fn
  
−
Counts for individual lexicons:
+
Precision: 100.00000%
 +
 +
Recall: 65.92357%
  
−
Punctuation:
+
===Current Coverage===
 +
(Bible Corpus)
  
−
ClassPrefix:  
+
Coverage: 61.5%
  
−
NounStem:  
+
Remaining Unknown forms: 803
  
−
PresentContinuousCircumfix:  
+
Ambiguity: 1.0880 (after disambiguation)
  
−
SubjectPrefix:
 
  
−
TenseSign:
+
(Narratives Corpus)
  
−
VerbRoot:  
+
Coverage: 52.2%
  
−
PositiveRecentPastCircumfix:  
+
Remaining Unknown forms: 468
  
−
NRPSubjectPrefix:  
+
Ambiguity: 1.1255 (after disambiguation)
  
−
NegativeRecentPastCircumfix:
 
  
−
PositiveImmediatePastCircumfix:
+
(Stories Corpus)
  
−
FUTSubjectPrefix:  
+
Coverage: 54.6%
  
−
Suffix:  
+
Remaining Unknown forms: 34
  
−
Numeral:  
+
Ambiguity: 1.1622 (after disambiguation)
  
−
All anonymous lexicons:
+
===Number of Stems in the Transducer===
  
 
===Top Unknown Words===
 
===Top Unknown Words===
 +
(Across Corpora)
  
 +
    14 ^ninche/*ninche$
 +
    13 ^ina/*ina$
 +
    13 ^Daudi/*Daudi$
 +
    11 ^lelo/*lelo$
 +
    10 ^tungʼana/*tungʼana$
 +
    10 ^Nkai/*Nkai$
 +
      9 ^Eliya/*Eliya$
 +
      8 ^to/*to$
 +
      8 ^ltungʼana/*ltungʼana$
 +
      8 ^lenyena/*lenyena$
 +
      6 ^aatookuo/*aatookuo$
 +
      5 ^lotimi/*lotimi$
 +
      5 ^ljipeta/*ljipeta$
 +
      5 ^Inno/*Inno$
 +
      4 ^p/*p$
 +
      4 ^nerinyunye/*nerinyunye$
 +
      4 ^neipiri/*neipiri$
 +
      4 ^nedol/*nedol$
 +
      4 ^naisho/*naisho$
 +
      4 ^lkak/*lkak$
 +
      2 ^nkaitin/*nkaitin$
 +
      2 ^keeya/*keeya$
 +
      2 ^Maayieu/*Maayieu$
 +
      2 ^Kaayieu/*Kaayieu$
 +
      1 ^tiapa/*tiapa$
 +
      1 ^tenkiterunoto/*tenkiterunoto$
 +
      1 ^papa/*papa$
 +
      1 ^olkurruk/*olkurruk$
 +
      1 ^neyam/*neyam$
 +
      1 ^nemeetae/*nemeetae$
  
 
===Analysis Tests===
 
===Analysis Tests===
−
 
  
 
==Generator Evaluation==
 
==Generator Evaluation==
Line 72: Line 182:
  
 
==Notes==
 
==Notes==
−
*  Number of twol rules:  
+
*  Number of twol rules: 8
−
*  
+
*
−
 
 
  
  
  
 
[[Category:Maasai]][[Category:sp23_FinalProjects]]
 
[[Category:Maasai]][[Category:sp23_FinalProjects]]

Latest revision as of 21:02, 7 May 2023

Transducer for Maasai

Transducer Repo: https://github.com/kdandy1/ling073-mas

Classes/Tenses Added

  • Nouns: nominative, accusative, vocative
  • Adjectives
  • Verbs: present, imperfect, past

Morphtests

Nouns

Note: all of the grammar books just say that there is no set rule to form the plural of nouns (or adjectives), so just memorize, so we're gonna do that too.

Nominative

Feminine: en prefix in the singular and in in the plural

  • tito<n><f><sg><nom> ↔ entito
  • tito<n><f><pl><nom> ↔ intoyie
  • kitok<n><f><sg><nom> ↔ enkitok
  • kitok<n><f><pl><nom> ↔ inkituaak
  • jore<n><f><sg><nom> ↔ enjore
  • jore<n><f><pl><nom> ↔ injorin
  • kima<n><f><sg><nom> ↔ enkima
  • kima<n><f><pl><nom> ↔ inkimaitie
  • aina<n><f><sg><nom> ↔ enkaina
  • aina<n><f><pl><nom> ↔ inkaik

Masculine: ol prefix in the singular and il in the plural

  • ayioni<n><m><sg><nom> ↔ olayioni
  • ayioni<n><m><pl><nom> ↔ ilayiok
  • payian<n><m><sg><nom> ↔ olpayian
  • payian<n><m><pl><nom> ↔ ilpayiani
  • lee<n><m><sg><nom> ↔ olee
  • lee<n><m><pl><nom> ↔ ilewa
  • soit<n><m><sg><nom> ↔ osoit
  • soit<n><m><pl><nom> ↔ isoito
  • reyiet<n><m><sg><nom> ↔ oreyiet
  • reyiet<n><m><pl><nom> ↔ ireyieta

Neuter: only one noun is neuter haha, prefixes are e in the singular and i in the plural

  • wueji<n><nt><pl><nom> ↔ ewueji
  • wueji<n><nt><pl><nom> ↔ iwuejitin

Adjectives

  • arriyia<adj><sg> ↔ arriyia
  • arriyia<adj><pl> ↔ arriyak
  • bioto<adj><sg> ↔ bioto
  • bioto<adj><pl> ↔ biot
  • dalut<adj><sg> ↔ dalut
  • dalut<adj><pl> ↔ daluti
  • asipani<adj><sg> ↔ asipani
  • asipani<adj><pl> ↔ asipak
  • kitok<adj><sg> ↔ kitok
  • kitok<adj><pl> ↔ kituaak

Verbs

Present Tense

prefix + root "to follow"

  • suj<vblex><prs><p1><sg> ↔ asuj
  • suj<vblex><prs><p1><pl> ↔ kisuj
  • suj<vblex><prs><p2><sg> ↔ isuj
  • suj<vblex><prs><p2><pl> ↔ isujusuju
  • suj<vblex><prs><p3><sg> ↔ esuj
  • suj<vblex><prs><p3><pl> ↔ esuj

"to wash"

  • isuj<vblex><prs><p1><sg> ↔ aisuj
  • isuj<vblex><prs><p1><pl> ↔ kiisuj
  • isuj<vblex><prs><p2><sg> ↔ iisuj
  • isuj<vblex><prs><p2><pl> ↔ iisujuisuju
  • isuj<vblex><prs><p3><sg> ↔ eisuj
  • isuj<vblex><prs><p3><pl> ↔ eisuj

Imperfect Progressive

prefix + root + suffix "to follow"

  • suj<vblex><prs_impf><p1><sg> ↔ asujita
  • suj<vblex><prs_impf><p1><pl> ↔ kisujita
  • suj<vblex><prs_impf><p2><sg> ↔ isujita
  • suj<vblex><prs_impf><p2><pl> ↔ isujita
  • suj<vblex><prs_impf><p3><sg> ↔ esujita
  • suj<vblex><prs_impf><p3><pl> ↔ esujita

Past

a + (t vowel) + root + suffix

various changes based on root vowel being open and closed

  • suj<vblex><past><p1><sg> ↔ atusuja
  • suj<vblex><past><p1><pl> ↔ kitusuja
  • suj<vblex><past><p2><sg> ↔ itusuja
  • suj<vblex><past><p2><pl> ↔ itusuja
  • suj<vblex><past><p3><sg> ↔ etusuja
  • suj<vblex><past><p3><pl> ↔ etusuja

Narrative

n + vowel + present tense form

  • suj<vblex><narr><p1><sg> ↔ nasuj
  • suj<vblex><narr><p1><pl> ↔ nikisuj
  • suj<vblex><narr><p2><sg> ↔ nisuj
  • suj<vblex><narr><p2><pl> ↔ nisujusuju
  • suj<vblex><narr><p3><sg> ↔ nesuj
  • suj<vblex><narr><p3><pl> ↔ nesuj

Analyser Evaluation

Precision and Recall

Totals: 152 forms, 199 tp, 8 fp, 0 tn, 111 fn

Precision: 100.00000%

Recall: 65.92357%

Current Coverage

(Bible Corpus)

Coverage: 61.5%

Remaining Unknown forms: 803

Ambiguity: 1.0880 (after disambiguation)


(Narratives Corpus)

Coverage: 52.2%

Remaining Unknown forms: 468

Ambiguity: 1.1255 (after disambiguation)


(Stories Corpus)

Coverage: 54.6%

Remaining Unknown forms: 34

Ambiguity: 1.1622 (after disambiguation)

Number of Stems in the Transducer

Top Unknown Words

(Across Corpora)

    14 ^ninche/*ninche$
    13 ^ina/*ina$
    13 ^Daudi/*Daudi$
    11 ^lelo/*lelo$
    10 ^tungʼana/*tungʼana$
    10 ^Nkai/*Nkai$
     9 ^Eliya/*Eliya$
     8 ^to/*to$
     8 ^ltungʼana/*ltungʼana$
     8 ^lenyena/*lenyena$
     6 ^aatookuo/*aatookuo$
     5 ^lotimi/*lotimi$
     5 ^ljipeta/*ljipeta$
     5 ^Inno/*Inno$
     4 ^p/*p$
     4 ^nerinyunye/*nerinyunye$
     4 ^neipiri/*neipiri$
     4 ^nedol/*nedol$
     4 ^naisho/*naisho$
     4 ^lkak/*lkak$
     2 ^nkaitin/*nkaitin$
     2 ^keeya/*keeya$
     2 ^Maayieu/*Maayieu$
     2 ^Kaayieu/*Kaayieu$
     1 ^tiapa/*tiapa$
     1 ^tenkiterunoto/*tenkiterunoto$
     1 ^papa/*papa$
     1 ^olkurruk/*olkurruk$
     1 ^neyam/*neyam$
     1 ^nemeetae/*nemeetae$

Analysis Tests

Generator Evaluation

Generator Tests

Notes

  • Number of twol rules: 8