Modeling Morphosyntactic Agreement in Constituency-Based Parsing of Modern Hebrew

Reut Tsarfaty; Khalil Sima'An

Modeling Morphosyntactic Agreement in Constituency-Based Parsing of Modern Hebrew

Reut Tsarfaty, Khalil Sima'An

Research output: Contribution to conference › Paper › peer-review

Abstract

We show that naïve modeling of morphosyntactic agreement in a Constituency-Based (CB) statistical parsing model is worse than none, whereas a linguistically adequate way of modeling inflectional morphology in CB parsing leads to improved performance. In particular, we show that an extension of the Relational-Realizational (RR) model that incorporates agreement features is superior to CB models that treat morphosyntax as state-splits (SP), and that the RR model benefits more from inflectional features. We focus on parsing Hebrew and report the best result to date, F₁84.13 for parsing off of gold-tagged text, 5% error reduction from previous results.

Original language	English
Pages	40-48
Number of pages	9
State	Published - 2010
Externally published	Yes
Event	1st Workshop on Statistical Parsing of Morphologically-Rich Languages, SPMRL 2010 - Los Angeles, United States Duration: 5 Jun 2010 → …

Conference

Conference	1st Workshop on Statistical Parsing of Morphologically-Rich Languages, SPMRL 2010
Country/Territory	United States
City	Los Angeles
Period	5/06/10 → …

Bibliographical note

Publisher Copyright:
© 2010 Association for Computational Linguistics

Cite this

@conference{ae06b534c13b49258218399fae2f81b4,

title = "Modeling Morphosyntactic Agreement in Constituency-Based Parsing of Modern Hebrew",

abstract = "We show that na{\"i}ve modeling of morphosyntactic agreement in a Constituency-Based (CB) statistical parsing model is worse than none, whereas a linguistically adequate way of modeling inflectional morphology in CB parsing leads to improved performance. In particular, we show that an extension of the Relational-Realizational (RR) model that incorporates agreement features is superior to CB models that treat morphosyntax as state-splits (SP), and that the RR model benefits more from inflectional features. We focus on parsing Hebrew and report the best result to date, F184.13 for parsing off of gold-tagged text, 5% error reduction from previous results.",

author = "Reut Tsarfaty and Khalil Sima'An",

note = "Publisher Copyright: {\textcopyright} 2010 Association for Computational Linguistics; 1st Workshop on Statistical Parsing of Morphologically-Rich Languages, SPMRL 2010 ; Conference date: 05-06-2010",

year = "2010",

language = "אנגלית",

pages = "40--48",

}

TY - CONF

T1 - Modeling Morphosyntactic Agreement in Constituency-Based Parsing of Modern Hebrew

AU - Tsarfaty, Reut

AU - Sima'An, Khalil

PY - 2010

Y1 - 2010

N2 - We show that naïve modeling of morphosyntactic agreement in a Constituency-Based (CB) statistical parsing model is worse than none, whereas a linguistically adequate way of modeling inflectional morphology in CB parsing leads to improved performance. In particular, we show that an extension of the Relational-Realizational (RR) model that incorporates agreement features is superior to CB models that treat morphosyntax as state-splits (SP), and that the RR model benefits more from inflectional features. We focus on parsing Hebrew and report the best result to date, F184.13 for parsing off of gold-tagged text, 5% error reduction from previous results.

AB - We show that naïve modeling of morphosyntactic agreement in a Constituency-Based (CB) statistical parsing model is worse than none, whereas a linguistically adequate way of modeling inflectional morphology in CB parsing leads to improved performance. In particular, we show that an extension of the Relational-Realizational (RR) model that incorporates agreement features is superior to CB models that treat morphosyntax as state-splits (SP), and that the RR model benefits more from inflectional features. We focus on parsing Hebrew and report the best result to date, F184.13 for parsing off of gold-tagged text, 5% error reduction from previous results.

UR - http://www.scopus.com/inward/record.url?scp=84874635173&partnerID=8YFLogxK

M3 - ???researchoutput.researchoutputtypes.contributiontoconference.paper???

AN - SCOPUS:84874635173

SP - 40

EP - 48

T2 - 1st Workshop on Statistical Parsing of Morphologically-Rich Languages, SPMRL 2010

Y2 - 5 June 2010

ER -

Modeling Morphosyntactic Agreement in Constituency-Based Parsing of Modern Hebrew

Abstract

Conference

Bibliographical note

Other files and links

Fingerprint

Cite this