Konferenzbeitrag

Hard constraints for grammatical function labelling

For languages with (semi-) free word order (such as German), labelling grammatical functions on top of phrase-structural constituent analyses is crucial for making them interpretable. Unfortunately, most statistical classifiers consider only local information for function labelling and fail to capture important restrictions on the distribution of core argument functions such as subject, object etc., namely that there is at most one subject (etc.) per clause. We augment a statistical classifier with an integer linear program imposing hard linguistic constraints on the solution space output by the classifier, capturing global distributional restrictions. We show that this improves labelling quality, in particular for argument grammatical functions, in an intrinsic evaluation, and, importantly, grammar coverage for treebankbased (Lexical-Functional) grammar acquisition and parsing, in an extrinsic evaluation.

Hard constraints for grammatical function labelling

Urheber*in: Seeker, Wolfgang; Rehbein, Ines; Kuhn, Joans; van Genabith, Josef

Urheberrechtsschutz

0
/
0

Sprache
Englisch

Thema
Phrasenstruktur
Automatische Sprachanalyse
Sprache

Ereignis
Geistige Schöpfung
(wer)
Seeker, Wolfgang
Rehbein, Ines
Kuhn, Joans
van Genabith, Josef
Ereignis
Veröffentlichung
(wer)
Stroudsburg, PA : Association for Computational Linguistics
(wann)
2016-11-21

URN
urn:nbn:de:bsz:mh39-56059
Letzte Aktualisierung
06.03.2025, 09:00 MEZ

Datenpartner

Dieses Objekt wird bereitgestellt von:
Leibniz-Institut für Deutsche Sprache - Bibliothek. Bei Fragen zum Objekt wenden Sie sich bitte an den Datenpartner.

Objekttyp

  • Konferenzbeitrag

Beteiligte

  • Seeker, Wolfgang
  • Rehbein, Ines
  • Kuhn, Joans
  • van Genabith, Josef
  • Stroudsburg, PA : Association for Computational Linguistics

Entstanden

  • 2016-11-21

Ähnliche Objekte (12)