Usi Using Pa PaQu for or l language acquis isit ition ion r - - PowerPoint PPT Presentation

▶
usi using pa paqu for or l language acquis isit ition ion
SMART_READER_LITE
LIVE PREVIEW

Usi Using Pa PaQu for or l language acquis isit ition ion r - - PowerPoint PPT Presentation

May 31, 2023 469 likes •762 views

Usi Using Pa PaQu for or l language acquis isit ition ion r research Jan Odijk CLARIN 2015 Conference Wroclaw, 2015-10-16 1 Overview Introduction CHILDES Corpora PaQu Evaluation & Analysis Conclusions Future

slide-1
SLIDE 1

Usi Using Pa PaQu for

  • r l

language acquis isit ition ion r research

Jan Odijk CLARIN 2015 Conference Wroclaw, 2015-10-16

1

slide-2
SLIDE 2
  • Introduction
  • CHILDES Corpora
  • PaQu
  • Evaluation & Analysis
  • Conclusions
  • Future Work

Overview

2

slide-3
SLIDE 3

(See [Odijk 2011, 2014] for more data and qualifications

Introduction

3 Cat init modifier predicate rest

A Hij is daar Heel / erg /zeer blij mee gloss He is there very happy with P Hij is daar *heel / erg / zeer in zijn sas mee gloss He is there very happy with V …omdat dat mij *heel / erg / zeer verbaast gloss …because that me very surprises

slide-4
SLIDE 4
  • Distinction is purely syntactic
  • Cannot be derived from semantic differences
  • Correlation with other known facts unlikely
  • Cannot be derived from general (universal)

principles

  •  must be acquired by L1 learners of Dutch

Introduction

4

slide-5
SLIDE 5
  • Minimal pair in acquisition
  • Requires acquisition of negative property

– No evidence in the input – No ‘correction’ or correction ignored

  • May provide evidence for/against relevant

hypotheses

– E.g. Indirect Negative Evidence hypothesis

  • Absence of evidence  evidence for absence

Introduction

5

slide-6
SLIDE 6
  • Problem: Ambiguity

– Heel 7-fold ambiguous – Erg 4-fold ambiguous – Zeer 3-fold ambiguous

  • (as any decent natural language word)
  • For our purposes:

– Morpho-syntactic and syntactic properties resolve the ambuigities

Corpus Analysis

6

slide-7
SLIDE 7
  • [Odijk 2014]
  • Automatic Corpus analysis: GrETEL, OpenSONAR, COAVA ,

LWRS, CMD

  • These apply to specific corpora only
  • Manual Corpus analysis of CHILDES Van Kampen Corpus
  • How can I apply these applications to my own corpus?
  •  request for PaQu (extends LWRS), AutoSearch (extends

CMD), …

Corpus Analysis

7

slide-8
SLIDE 8
  • PaQu= Parse and Query: https://dev.clarin.nl/node/4182
  • Web application made by Groningen University
  • Upload corpus

– Plain text or in Alpino format

  • Plain Text is automatically parsed by Alpino
  • Resulting treebank can be searched and analyzed

– Search

  • Word relations interface and XPATH Queries

– Analysis

  • User-definable statistics on search results (and metadata)

PaQu

8

slide-9
SLIDE 9
  • Take the Dutch CHILDES corpora
  • Select all utterances containing heel, erg or zeer
  • Clean the utterances, e.g.
  • ja , maar <we be> [//] we bewaren (he)t ook
  • ja , maar we bewaren het ook
  • Upload it into PaQu
  • Gather statistics and draw conclusions

Experiments

9

slide-10
SLIDE 10
  • Adult utterances of Van Kampen Corpus
  • Manual annotation used as gold standard (Acc)
  • Alpino makes finer distinctions: I mapped these
  • Annotation errors in the gold standard: revised gold

standard (Rev Acc)

Experiment 1

10

slide-11
SLIDE 11
  • Accuracy

Experiment 1: Results

11

word Acc Rev Acc heel 0.94 0.95 erg 0.88 0.91 zeer 0.21 0.21

slide-12
SLIDE 12
  • Good for heel, erg
  • Bad for zeer, but:
  • Completely due to zeer doen (lit. pain(ful) do, ‘to hurt’)
  • Can be identified very easily in PaQu
  • Generalisability: Limited
  • It concerns (cleaned) adult speech
  • It concerns relatively short sentences, explicitly separated
  • It mostly concerns a very local grammatical relation

Experiment 1: Interpretation

12

slide-13
SLIDE 13
  • All adults’ utterances:

Experiment 2:

13

Results mod A mod N Mod V mod P predc other unclear Total heel 886 46 2 2 14 2 952 erg 347 27 109 187 5 675 zeer 7 1 83 19 21 7 138

slide-14
SLIDE 14
  • Heel most frequent (almost 54%)

Experiment 2: Interpretation

14

Results mod A mod N Mod V mod P predc other unclear Total heel 886 46 2 2 14 2 952 erg 347 27 109 187 5 675 zeer 7 1 83 19 21 7 138

slide-15
SLIDE 15
  • Heel as mod A overwhelming: > 93%

Experiment 2: Interpretation

15

Results mod A mod N Mod V mod P predc other unclear Total heel 886 46 2 2 14 2 952 erg 347 27 109 187 5 675 zeer 7 1 83 19 21 7 138

slide-16
SLIDE 16
  • Heel as mod V, mod P wrong analysis

Experiment 2: Interpretation

16

Results mod A mod N Mod V mod P predc other unclear Total heel 886 46 2 2 14 2 952 erg 347 27 109 187 5 675 zeer 7 1 83 19 21 7 138

slide-17
SLIDE 17
  • Mod A and mod V more balanced for erg

Experiment 2: Interpretation

17

Results mod A mod N Mod V mod P predc other unclear Total heel 886 46 2 2 14 2 952 erg 347 27 109 187 5 675 zeer 7 1 83 19 21 7 138

slide-18
SLIDE 18
  • Evidence for zeer mostly lacking
  • Cases of Mod V are mostly wrong analyses

Experiment 2: Interpretation

18

Results mod A mod N Mod V mod P predc other unclear Total heel 886 46 2 2 14 2 952 erg 347 27 109 187 5 675 zeer 7 1 83 19 21 7 138

slide-19
SLIDE 19
  • Evidence for Mod P mostly lacking
  • Some evidence for erg, zeer (4 occurrences)

Experiment 2: Interpretation

19

Results mod A mod N Mod V mod P predc other unclear Total heel 886 46 2 2 14 2 952 erg 347 27 109 187 5 675 zeer 7 1 83 19 21 7 138

slide-20
SLIDE 20
  • Van Kampen Children’s speech: Accuracy
  • Similar to the Adults’ speech but slightly lower

Experiment 3:

20

Word Acc heel 0.90 erg 0.73 zeer 0.17

slide-21
SLIDE 21
  • Linguistics:
  • No examples for mod P: how to explain heel v. erg, zeer?
  • Overwhelmingness of mod A for heel might be a relevant

factor

  • Current Dutch CHILDES corpora probably too small to draw

reliable conclusions

Conclusions

21

slide-22
SLIDE 22
  • PaQu:
  • PaQu is very useful for doing better and more efficient

manual verification of hypotheses

  • In some cases its parses and their statistics can reliably be

used directly (though care is required!)

  • Several small details were improved, small additions to

functionality made through these experiments

Conclusions

22

slide-23
SLIDE 23
  • More experiments for the children’s speech (cf. [Odijk

2014:34])

  • Similar experiments for other examples
  • te ‘too’ v. overmatig ‘excessively’; worden ‘become’v. raken ‘get’

and others

  • Extend PaQu to include all relevant `metadata’
  • Extend PaQu to natively support common formats such as

CHAT, Folia, TEI, …

  • Make similar system for GrETEL, OpenSONAR
  • Manually verify (parts of) parses for CHILDES corpora

(most is being done in CLARIAH-NL or UU AnnCor)

Future Work

23

slide-24
SLIDE 24

Thanks for Attention!

Visit the Demo at 16:30! Visit the Bazaar at 14:30 for a completely different use of PaQu!

24

slide-25
SLIDE 25

 NO!

Correlation with other Differences?

25

Phenomenon Opposes Versus Mod V,P heel erg, zeer Meaning erg heel, zeer Inflection heel, erg zeer Comparative, Superlative erg heel, zeer Modifiee erg heel, zeer Pragmatics zeer heel, erg

slide-26
SLIDE 26

Ambiguity: HEEL

26

word Morpho- syntax Syntax Meaning heel A Mod N (1)`whole’ (2) ‘in one piece’ (3)`large’ Predc ‘in one piece’ Mod A `very’ Vf (1)`heal’ (2) `receive’

slide-27
SLIDE 27

Ambiguity: ERG

27

word Morpho- syntax Syntax Meaning erg N utrum `erg’ N neutrum `evil’ A Mod N, predc ‘bad’, ‘awful’ Mod A V P very

slide-28
SLIDE 28

Ambiguity: ZEER

28

word Morpho- Syntax Syntax Meaning zeer N `pain’ A Mod N, predc ‘painful’ Mod A V P ‘very’