Mova0.5 pre-release⌕
☰
home / corpora / questions / AI / Cars, doctors, scientists and translat… / t1

Test 1 · Cars, doctors, scientists and translators

"translator app did my nan's letter from the council, its fine yeah?" Best reply?

Mova's choice: option 3 · The right answer: option 3 · right ✓ · Who acts: translator app, nan · What happens: do, letter · To what, where: letter, council

This is test 1 of lesson 18 «Cars, doctors, scientists and translators» of the AI course of Mova's primary school. The page gives the question's syntax tree — a dependency tree in the style of Universal Dependencies — and its pragmatics clause by clause, then the 4 options as the student saw them, each with its tree and pragmatics. Every text is parsed by Mova itself. How the student answered it, with the score of every option and why: the test at school.

The question

Sentence 1: "translator app did my nan's letter from the council, its fine yeah?"

punctcompoundnsubjnmod:possnmod:posscaseobjcasedetnmodpunctnmod:possconjdiscoursepunctpunctroot"PUNCTtranslatorNOUNappNOUNdidVERBmyPRONnanNOUN'sPARTletterNOUNfromADPtheDETcouncilNOUN,PUNCTitsPRONfineADJyeahINTJ?PUNCT"PUNCT
The words, one row each (CoNLL-U)
#wordlemmaUPOSfeaturesheadrelation
1""PUNCT_4 didpunct
2translatortranslatorNOUNNumber=Sing3 appcompound
3appappNOUNNumber=Sing4 didnsubj
4diddoVERBMood=Ind Number=Sing Person=3 Tense=Past VerbForm=Finrootroot
5mymyPRONCase=Gen Number=Sing Person=1 Poss=Yes PronType=Prs6 nannmod:poss
6nannanNOUNNumber=Sing8 letternmod:poss
7's'sPART_6 nancase
8letterletterNOUNNumber=Sing4 didobj
9fromfromADP_11 councilcase
10thetheDETDefinite=Def PronType=Art11 councildet
11councilcouncilNOUNNumber=Sing8 letternmod
12,,PUNCT_14 finepunct
13itsitsPRONCase=Gen Gender=Neut Number=Sing Person=3 Poss=Yes PronType=Prs14 finenmod:poss
14finefineADJDegree=Pos4 didconj
15yeahyeahINTJ_14 finediscourse
16??PUNCT_4 didpunct
17""PUNCT_4 didpunct
#verbtense, moodwhowhat, whomto whomwherewhenotherroles by
1do in a quotepasttranslator appmy nan's letterthe councilvn act-114-1 «NP V NP»: agent=Agent letter=Eventuality
2letter in a quotemy nanthe councilnom letter «communication transitive»: poss=agent from=place

One row per clause the reader finds. «not said here»: a role filled from the clause above. «roles by»: a VerbNet class and frame when one fits the verb, else the reader's rules on the tree.

Sentence 2: Best reply?

amodpunctrootBestADJreplyNOUN?PUNCT
The words, one row each (CoNLL-U)
#wordlemmaUPOSfeaturesheadrelation
1BestgoodADJDegree=Sup2 replyamod
2replyreplyNOUNNumber=Singrootroot
3??PUNCT_2 replypunct

The frames reader finds no event or fact in this sentence.

The options

Option 1

Yes, councils write in robot language anyway

discoursepunctnsubjcasecompoundobladvmodrootYesINTJ,PUNCTcouncilsNOUNwriteVERBinADProbotNOUNlanguageNOUNanywayADV
The words, one row each (CoNLL-U)
#wordlemmaUPOSfeaturesheadrelation
1YesyesINTJPolarity=Pos4 writediscourse
2,,PUNCT_1 Yespunct
3councilscouncilNOUNNumber=Plur4 writensubj
4writewriteVERBMood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Finrootroot
5ininADP_7 languagecase
6robotrobotNOUNNumber=Sing7 languagecompound
7languagelanguageNOUNNumber=Sing4 writeobl
8anywayanywayADV_4 writeadvmod
#verbtense, moodwhowhat, whomto whomwherewhenotherroles by
1writepresentcouncilsrobot languageanyway propertyvn scribble-25.2-1 «NP V NP PP.destination»: agent=Agent in language=Destination

Option 2

Yes, translator apps never make mistakes

discoursepunctcompoundnsubjadvmodobjrootYesINTJ,PUNCTtranslatorNOUNappsNOUNneverADVmakeVERBmistakesNOUN
The words, one row each (CoNLL-U)
#wordlemmaUPOSfeaturesheadrelation
1YesyesINTJPolarity=Pos4 appsdiscourse
2,,PUNCT_1 Yespunct
3translatortranslatorNOUNNumber=Sing4 appscompound
4appsappNOUNNumber=Plur6 makensubj
5neverneverADVPronType=Neg6 makeadvmod
6makemakeVERBMood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Finrootroot
7mistakesmistakeNOUNNumber=Plur6 makeobj
#verbtense, moodwhowhat, whomto whomwherewhenotherroles by
1not makepresenttranslator appsmistakesvn build-26.1-1 «NP.asset V NP»: agent=Asset mistake=Product

Option 3 · ← Mova's choice · ✓ the right answer

For something important, get a person who speaks both languages to check

caseoblamodpunctdetobjnsubjacl:relcldetobjmarkadvclrootForADPsomethingPRONimportantADJ,PUNCTgetVERBaDETpersonNOUNwhoPRONspeaksVERBbothDETlanguagesNOUNtoPARTcheckVERB
The words, one row each (CoNLL-U)
#wordlemmaUPOSfeaturesheadrelation
1ForforADP_2 somethingcase
2somethingsomethingPRONNumber=Sing PronType=Ind5 getobl
3importantimportantADJDegree=Pos2 somethingamod
4,,PUNCT_2 somethingpunct
5getgetVERBMood=Imp VerbForm=Finrootroot
6aaDETDefinite=Ind PronType=Art7 persondet
7personpersonNOUNNumber=Sing5 getobj
8whowhoPRONPronType=Rel9 speaksnsubj
9speaksspeakVERBMood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin7 personacl:relcl
10bothbothDETPronType=Tot11 languagesdet
11languageslanguageNOUNNumber=Plur9 speaksobj
12totoPART_13 checkmark
13checkcheckVERBVerbForm=Inf9 speaksadvcl
#verbtense, moodwhowhat, whomto whomwherewhenotherroles by
1getimperativeyou not said herea personfor somethingvn get-13.5.1-1 «NP V NP PP.source»: agent=Agent person=Theme for something=Source
2speak speech: otherpresenta personboth languagestree rules
3checkto-infinitivea person not said herevn assessment-34.1 «NP V NP»: agent=Agent

Option 4

No, translator apps only work for short words, not for whole letters

discoursepunctcompoundnsubjadvmodcaseamodoblpunctadvmodcaseamodoblrootNoINTJ,PUNCTtranslatorNOUNappsNOUNonlyADVworkVERBforADPshortADJwordsNOUN,PUNCTnotPARTforADPwholeADJlettersNOUN
The words, one row each (CoNLL-U)
#wordlemmaUPOSfeaturesheadrelation
1NonoINTJPolarity=Neg4 appsdiscourse
2,,PUNCT_1 Nopunct
3translatortranslatorNOUNNumber=Sing4 appscompound
4appsappNOUNNumber=Plur6 worknsubj
5onlyonlyADV_6 workadvmod
6workworkVERBMood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Finrootroot
7forforADP_9 wordscase
8shortshortADJDegree=Pos9 wordsamod
9wordswordNOUNNumber=Plur6 workobl
10,,PUNCT_14 letterspunct
11notnotPARTPolarity=Neg14 lettersadvmod
12forforADP_14 letterscase
13wholewholeADJDegree=Pos14 lettersamod
14lettersletterNOUNNumber=Plur6 workobl
#verbtense, moodwhowhat, whomto whomwherewhenotherroles by
1workpresenttranslator appsfor short words; for whole lettersvn employment-95.3 «NP V PP.beneficiary»: agent=Agent for word=Beneficiary

Text: the tests of Mova's primary school, written for Mova (world/data/school/), as the course runner gave them to the student. Trees: Mova's own parse — the en model on the CPU, when the site is built — not checked by a teacher; they may be wrong. Frames: Mova's frames reader (prag), each text read on its own. Elamur Cosmic License.

This page in Markdown → https://md.mova.elamur.ai/free/en/corpora/questions/ai/18-cars-doctors-translators/t1/