Argumentative texts and clause types Alexis Palmer Leibniz - - PowerPoint PPT Presentation

▶
argumentative texts and clause types
SMART_READER_LITE
LIVE PREVIEW

Argumentative texts and clause types Alexis Palmer Leibniz - - PowerPoint PPT Presentation

Mar 11, 2024 168 likes •500 views

Argumentative texts and clause types Alexis Palmer Leibniz ScienceCampus, University of Heidelberg, IDS (Mannheim) Dagstuhl, April 2016 1 Carlota Smith Argumentative texts and clause types Annemarie Friedrich, Saarland University Alexis

slide-1
SLIDE 1

Argumentative texts and clause types

Alexis Palmer

Leibniz ScienceCampus, University of Heidelberg, IDS (Mannheim)

Dagstuhl, April 2016

1

slide-2
SLIDE 2

Argumentative texts and clause types

Alexis Palmer

Leibniz ScienceCampus, University of Heidelberg, IDS (Mannheim)

Dagstuhl, April 2016

2

Annemarie Friedrich, Saarland University Carlota Smith

slide-3
SLIDE 3

Argumentative texts and clause types

Alexis Palmer

Leibniz ScienceCampus, University of Heidelberg, IDS (Mannheim)

Dagstuhl, April 2016

3

Maria Becker Anette Frank

slide-4
SLIDE 4

Alexis Palmer Dagstuhl, April 2016

Situation entities

  • Overview: new project annotating argumentative

microtexts (Peldszus & Stede) with SE types

  • Situation entity (SE) type — what kind of situation does

the clause evoke in the discourse?

  • SE type linked to linguistic characteristics of clause
  • lexical aspect of main verb
  • genericity of main referent (~ subject)
  • habituality of clause

4

slide-5
SLIDE 5

Alexis Palmer Dagstuhl, April 2016

Situation entities

  • Overview: new project annotating argumentative

microtexts (Peldszus & Stede) with SE types

  • Situation entity (SE) type — what kind of situation does

the clause evoke in the discourse?

  • SE type linked to linguistic characteristics of clause
  • lexical aspect of main verb
  • genericity of main referent (~ subject)
  • habituality of clause
  • … and can be modeled computationally and predicted

automatically

5

[Friedrich/Palmer, ACL14], 66-93% accuracy [Friedrich/Pinkal, ACL15], 70-89% accuracy [Friedrich/Pinkal, EMNLP15], 74-84% accuracy [Friedrich/Palmer/Pinkal, in submission], ~80% accuracy

slide-6
SLIDE 6

Alexis Palmer Dagstuhl, April 2016

Eventualities - states and events

(1) The NMFS estimates (2) that in 2000, more than 16,000 tons of Chilean seabass were legally caught from an internationally regulated harvest area in the Antarctic Ocean. (3) But more than 32,000 tons may have been taken illegally from those same waters, (4) the fisheries service said.

state event state* reporting

[masc_news_NYTnewswire7_part1.txt]

6

ST

slide-7
SLIDE 7

Alexis Palmer Dagstuhl, April 2016

Eventualities - states and events

(1) The NMFS estimates (2) that in 2000, more than 16,000 tons of Chilean seabass were legally caught from an internationally regulated harvest area in the Antarctic Ocean. (3) But more than 32,000 tons may have been taken illegally from those same waters, (4) the fisheries service said.

state event state* reporting

[masc_news_NYTnewswire7_part1.txt]

7

EV

slide-8
SLIDE 8

Alexis Palmer Dagstuhl, April 2016

Eventualities - states and events

(1) The NMFS estimates (2) that in 2000, more than 16,000 tons of Chilean seabass were legally caught from an internationally regulated harvest area in the Antarctic Ocean. (3) But more than 32,000 tons may have been taken illegally from those same waters, (4) the fisheries service said.

state event state* reporting

[masc_news_NYTnewswire7_part1.txt]

8

ST

slide-9
SLIDE 9

Alexis Palmer Dagstuhl, April 2016

Eventualities - states and events

(1) The NMFS estimates (2) that in 2000, more than 16,000 tons of Chilean seabass were legally caught from an internationally regulated harvest area in the Antarctic Ocean. (3) But more than 32,000 tons may have been taken illegally from those same waters, (4) the fisheries service said.

state event state* reporting

[masc_news_NYTnewswire7_part1.txt]

9

REP

slide-10
SLIDE 10

Alexis Palmer Dagstuhl, April 2016

General Statives

(5) Blobfish are small fish, typically shorter than 30 cm. (6) Blobfish are often caught as bycatch in bottom trawling nets.

generic generalizing sentence (habitual)

[adapted from wikipedia_wikiGenerics_blobfish.txt]

10

G

slide-11
SLIDE 11

Alexis Palmer Dagstuhl, April 2016

General Statives

(5) Blobfish are small fish, typically shorter than 30 cm. (6) (?) Blobfish are often caught as bycatch in bottom trawling nets.

generic generalizing sentence (habitual)

[adapted from wikipedia_wikiGenerics_blobfish.txt]

11

GS

slide-12
SLIDE 12

Alexis Palmer Dagstuhl, April 2016

General Statives

(5) Blobfish are small fish, typically shorter than 30 cm. (6) (?) Blobfish are often caught as bycatch in bottom trawling nets. I have often caught blobfish accidentally on my fishing trips.

generic generalizing sentence (habitual)

[adapted from wikipedia_wikiGenerics_blobfish.txt]

12

GS

slide-13
SLIDE 13

Alexis Palmer Dagstuhl, April 2016

Abstract Entities

(7a) There is no doubt [state] (7b) that [Randall enjoys her work]S. (8a) Mr. Icahn then proposed (8b) that [USAir buy TWA].

fact proposition

[masc_journal_VOL15_3_part16.txt]

13

[MUC6 data]

F

slide-14
SLIDE 14

Alexis Palmer Dagstuhl, April 2016

Abstract Entities

fact proposition

(7a) There is no doubt (7b) that [Randall enjoys her work]S. (8a) Mr. Icahn then proposed [event] (8b) that [USAir buy TWA]S.

[masc_journal_VOL15_3_part16.txt]

14

[MUC6 data]

P

slide-15
SLIDE 15

Alexis Palmer Dagstuhl, April 2016

Abstract Entities

fact proposition

(7a) There is no doubt (7b) that [Randall enjoys her work]S. (8a) Mr. Icahn then proposed [event] (8b) that [USAir buy TWA]S.

[masc_journal_VOL15_3_part16.txt]

15

[MUC6 data]

PLUS: question

imperative

slide-16
SLIDE 16

Alexis Palmer Dagstuhl, April 2016

Abstract Entities

fact proposition

(7a) There is no doubt (7b) that [Randall enjoys her work]S. (8a) Mr. Icahn then proposed [event] (8b) that [USAir buy TWA]S.

[masc_journal_VOL15_3_part16.txt]

16

[MUC6 data]

PLUS: question

imperative

wait, but why?

slide-17
SLIDE 17

Alexis Palmer Dagstuhl, April 2016

Discourse modes

NARRATIVE( REPORT( INFORMATION( DESCRIPTION(

ARGUMENT/( COMMENTARY(

[Smith 2003]

Related: Werlich’s (1975) typology of texts, Santini (2006), i.a.

Discourse Modes

[images thanks to Kleio Mavridou]

17

slide-18
SLIDE 18

Alexis Palmer Dagstuhl, April 2016

Patterning of SEs and Genre

  • Taking genre categories as proxy for DMs, do SE types

pattern as the theory predicts?

  • Manual SE annotations
  • Manually-Annotated SubCorpus (MASC): 10,270 clauses

from news, jokes, essays, and (fund-raising) letters

  • Penn Discourse TreeBank (PDTB): 2513 clauses from

news and essays (following Webber 2009)

18

slide-19
SLIDE 19

Alexis Palmer Dagstuhl, April 2016

MASC: SEs and Genre

  • REPORT texts: greater proportion of States and Events,

fewer Generics and Generalizing Sentences

  • ARGUMENT/COMMENTARY texts: many more Generics

and Generalizing Sentences, States over Events

[Palmer/Friedrich 2014]

19

%

slide-20
SLIDE 20

Alexis Palmer Dagstuhl, April 2016

And now to focus on arguments

  • Argumentative microtexts offer corpus of “purely

argumentative” text passages

  • *with argument graphs*
  • Our question: How can SE types be helpful for modeling

argumentative regions of text? For mining arguments? For reasoning over argument components?

  • Annotation project underway - analysis of this data will

inform next steps

  • Next texts: from Potsdam Commentary Corpus

20

slide-21
SLIDE 21

Alexis Palmer Dagstuhl, April 2016

One example: micro_b005

21

[e1] Die Geheimdienste müssen dringend stärker vom Parlament kontrolliert werden, [e2] das sollte jedem nach den Enthüllungen von Edward Snowden klar sein.

1

[e3] Die betreffen zwar vor allem die britischen und amerikanischen Geheimdienste,

2

[e4] aber mit denen arbeiten die Deutschen Dienste bekanntermaßen eng zusammen.

3

[e5] Deren W erkzeuge, Daten und Knowhow wird schon lange zu unserer Überwachung genutzt.

4 5

c2 c5 c4 c3

[e1] Intelligence services must urgently be regulated more tightly by parliament; [e2] this should be clear to everyone after the disclosures

  • f Edward Snowden.

[e3] Granted, this concerns primarily the British and American intelligence services, [e4] but the German services evidently do collaborate with them closely. [e5] Their tools, data and expertise have been used to keep us under surveillance for a long time.

slide-22
SLIDE 22

Alexis Palmer Dagstuhl, April 2016

One example: micro_b005

22

[e1] Die Geheimdienste müssen dringend stärker vom Parlament kontrolliert werden, [e2] das sollte jedem nach den Enthüllungen von Edward Snowden klar sein.

1

[e3] Die betreffen zwar vor allem die britischen und amerikanischen Geheimdienste,

2

[e4] aber mit denen arbeiten die Deutschen Dienste bekanntermaßen eng zusammen.

3

[e5] Deren W erkzeuge, Daten und Knowhow wird schon lange zu unserer Überwachung genutzt.

4 5

c2 c5 c4 c3

G S S GS GS

[e1] Intelligence services must urgently be regulated more tightly by parliament; [e2] this should be clear to everyone after the disclosures

  • f Edward Snowden.

[e3] Granted, this concerns primarily the British and American intelligence services, [e4] but the German services evidently do collaborate with them closely. [e5] Their tools, data and expertise have been used to keep us under surveillance for a long time.

slide-23
SLIDE 23

Alexis Palmer Dagstuhl, April 2016

Preliminary (very) observations

  • Arg. texts are different from non-arg. texts wrt SE types
  • very few events
  • many generics and generalizing sentences
  • Claims tend to be generics or states
  • when generic, always first segment
  • when state, usually not first segment (often preceded

by counter-claims)

  • Small tendency for counter-claims to be non-generic vs.

generic proponent claims

23

slide-24
SLIDE 24

Alexis Palmer Dagstuhl, April 2016

Looking ahead…

  • For what kinds of tasks could it be useful to have clauses

labeled with SE types?

  • Use SE types for discourse mode identification and

classification (in progress) - find argumentative regions of texts

24

slide-25
SLIDE 25

Alexis Palmer Dagstuhl, April 2016

Tools and data

  • Coming soon (May/June 2016): most of MASC annotated

for SE type, lexical aspect, genericity, and habituality

  • Coming soon (May/June 2016): SE labeler for English
  • WikiGenerics: data annotated for genericity at clause and

NP level

  • AspMASC, AspAmbig: data annotated for lexical aspect
  • http://www.coli.uni-saarland.de/projects/sitent/page.php

25

slide-26
SLIDE 26

Thank you!

26

slide-27
SLIDE 27

Alexis Palmer Dagstuhl, April 2016

REFERENCES

Alexis Palmer, Jonas Kuhn, and Carlota Smith. 2004. Utilization of Multiple Language Resources for Robust Grammar-Based Tense and Aspect Classification. LREC, Lisbon. Alexis Palmer, Elias Ponvert, Jason Baldridge, and Carlota Smith. 2007. A Sequence Model for Situation Entity Classification. ACL, Prague. Alexis Palmer and Caroline Sporleder. 2009 (abstract & presentation only). Situation Entities and Genre Distinctions in the Penn Discourse TreeBank. TLS2009. Austin. Annemarie Friedrich and Alexis Palmer. 2014. Automatic prediction of aspectual class of verbs in

  • context. ACL, Baltimore.

Annemarie Friedrich and Alexis Palmer. 2014. Situation entity annotation. The Linguistic Annotation Workshop, Dublin. Nicholas Asher. 1993. Reference to Abstract Objects in Discourse. Kluwer. Gregory Carlson and F. Jeffry Pelletier, eds. 1995. The Generic Book. University of Chicago Press. Steven Clark and James Curran. 2004. Parsing the WSJ using CCG and log-linear models. ACL2004. David Dowty. 1979. Word Meaning and Montague Grammar. Springer Classic Titles in Linguistics. Philip L. Peterson. 1997. Fact Proposition Event. Kluwer. Carlota S. Smith. 1991. The Parameter of Aspect. Kluwer. Carlota S. Smith. 2003. Modes of Discourse. Cambridge University Press. Zeno Vendler. 1967. Linguistics in Philosophy, chap. Verbs and Times. Cornell University Press. Henk Verkuyl. 1972. On the Compositional Nature of the Aspects. Reidel.

27

slide-28
SLIDE 28

Extra slides

28

slide-29
SLIDE 29

Alexis Palmer Dagstuhl, April 2016

General Statives

(5) Blobfish are small fish, typically shorter than 30 cm. (6) Blobfish are often caught as bycatch in bottom trawling nets.

generic generalizing sentence (habitual)

(7a) Texans wear cowboy hats. (7b) The Texan wears cowboy hats. (cf. The Texan wore a cowboy hat.)

[constructed]

29

GS G

slide-30
SLIDE 30

Alexis Palmer Dagstuhl, April 2016

General Statives

(5) Blobfish are small fish, typically shorter than 30 cm. (6) Blobfish are often caught as bycatch in bottom trawling nets.

generic generalizing sentence (habitual)

(7a) Texans wear cowboy hats. (7b) The Texan wears cowboy hats. (cf. The Texan wore a cowboy hat.)

[constructed]

30

GS G EV

slide-31
SLIDE 31

Alexis Palmer Dagstuhl, April 2016

Feature-driven annotation

lexical aspect of main verb

genericity of main referent

habituality lexical aspect of main verb

[generic] [specific]

habituality

[dynamic] [stative] [dynamic] [stative] [can’t decide]

habituality habituality General Stative

[habitual] [habitual]

Generic Generalizing State Event

[episodic] [habitual] [static] [episodic] [modality, negation, conditionality, perfect tense]

31

Mary runs every Tuesday.

slide-32
SLIDE 32

Alexis Palmer Dagstuhl, April 2016

Feature-driven annotation

lexical aspect of main verb

genericity of main referent

habituality lexical aspect of main verb

[generic] [specific]

habituality

[dynamic] [stative] [dynamic] [stative] [can’t decide]

habituality habituality General Stative

[habitual] [habitual]

Generic Generalizing State Event

[episodic] [habitual] [static] [episodic] [modality, negation, conditionality, perfect tense]

32

Mary runs every Tuesday.