A derivational model of discontinuous parsing

Mark-Jan Nederhof; Anssi Yli-Jyrä

doi:10.1016/j.ic.2020.104619

A derivational model of discontinuous parsing

Mark-Jan Nederhof, Anssi Yli-Jyrä

School of Computer Science

Research output: Contribution to journal › Article › peer-review

1 Citation (Scopus)

18 Downloads (Pure)

Abstract

The notion of latent-variable probabilistic context-free derivation of syntactic structures is enhanced to allow heads and unrestricted discontinuities. The chosen formalization covers both constituency parsing and dependency parsing. By the new framework, one obtains a probability distribution over the space of all discontinuous parses. This lends itself to intrinsic evaluation in terms of cross-entropy. The derivational model is accompanied by an equivalent automaton model, which can be used for deterministic parsing.

Original language	English
Article number	104619
Journal	Information and Computation
Volume	In press
Early online date	10 Aug 2020
DOIs	https://doi.org/10.1016/j.ic.2020.104619
Publication status	E-pub ahead of print - 10 Aug 2020

Keywords

Parsing
Grammars
Weighted automata

Access to Document

10.1016/j.ic.2020.104619

Nederhof-2020-Derivational-model-of-I&C-AAM
Copyright © 2020 Elsevier, Inc. This work has been made available online in accordance with publisher policies or with permission. Permission for further reuse of this content should be sought from the publisher or the rights holder. This is the author created accepted manuscript following peer review and may differ slightly from the final published version. The final published version of this work is available at https://doi.org/10.1016/j.ic.2020.104619.
Accepted author manuscript, 381 KB

Cite this

@article{4e55f91dc99c42a7a56e08eb597a7036,

title = "A derivational model of discontinuous parsing",

abstract = "The notion of latent-variable probabilistic context-free derivation of syntactic structures is enhanced to allow heads and unrestricted discontinuities. The chosen formalization covers both constituency parsing and dependency parsing. By the new framework, one obtains a probability distribution over the space of all discontinuous parses. This lends itself to intrinsic evaluation in terms of cross-entropy. The derivational model is accompanied by an equivalent automaton model, which can be used for deterministic parsing.",

keywords = "Parsing, Grammars, Weighted automata",

author = "Mark-Jan Nederhof and Anssi Yli-Jyr{\"a}",

year = "2020",

month = aug,

day = "10",

doi = "10.1016/j.ic.2020.104619",

language = "English",

volume = "In press",

journal = "Information and Computation",

issn = "0890-5401",

publisher = "Elsevier Inc.",

}

TY - JOUR

T1 - A derivational model of discontinuous parsing

AU - Nederhof, Mark-Jan

AU - Yli-Jyrä, Anssi

PY - 2020/8/10

Y1 - 2020/8/10

N2 - The notion of latent-variable probabilistic context-free derivation of syntactic structures is enhanced to allow heads and unrestricted discontinuities. The chosen formalization covers both constituency parsing and dependency parsing. By the new framework, one obtains a probability distribution over the space of all discontinuous parses. This lends itself to intrinsic evaluation in terms of cross-entropy. The derivational model is accompanied by an equivalent automaton model, which can be used for deterministic parsing.

AB - The notion of latent-variable probabilistic context-free derivation of syntactic structures is enhanced to allow heads and unrestricted discontinuities. The chosen formalization covers both constituency parsing and dependency parsing. By the new framework, one obtains a probability distribution over the space of all discontinuous parses. This lends itself to intrinsic evaluation in terms of cross-entropy. The derivational model is accompanied by an equivalent automaton model, which can be used for deterministic parsing.

KW - Parsing

KW - Grammars

KW - Weighted automata

U2 - 10.1016/j.ic.2020.104619

DO - 10.1016/j.ic.2020.104619

M3 - Article

SN - 0890-5401

VL - In press

JO - Information and Computation

JF - Information and Computation

M1 - 104619

ER -

A derivational model of discontinuous parsing

Abstract

Keywords

Access to Document

Fingerprint

Cite this