A Derivational Model of Discontinuous Parsing

Mark-Jan Nederhof; Anssi Mikael Yli-Jyrä

doi:10.1007/978-3-319-53733-7_22

A Derivational Model of Discontinuous Parsing

Mark-Jan Nederhof, Anssi Mikael Yli-Jyrä

Language Technology

Research output: Chapter in Book/Report/Conference proceeding › Chapter › Scientific › peer-review

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 constituent parsing and dependency parsing. The derivational model is accompanied by an equivalent probabilistic automaton model. 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 perplexity, as shown in experiments.

Translated title of the contribution	Epäjatkuvan jäsennyspuun johtoon perustuva tilastollinen malli
Original language	English
Title of host publication	Language and Automata Theory and Applications : LATA 2017
Editors	Carlos Martín-Vide
Number of pages	12
Place of Publication	Berlin
Publisher	Springer-Verlag
Publication date	6 Mar 2017
Pages	299-310
ISBN (Print)	978-3-319-53732-0
ISBN (Electronic)	978-3-319-53733-7
DOIs	https://doi.org/10.1007/978-3-319-53733-7_22
Publication status	Published - 6 Mar 2017
MoE publication type	A3 Book chapter

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Fields of Science

111 Mathematics
grammars
perplexity
113 Computer and information sciences
weighted automata
context-free derivation
discontinuity
derivation
6121 Languages
parsing
syntactic structures

Access to Document

10.1007/978-3-319-53733-7_22Licence: Unspecified

Cite this

@inbook{601967b67f9c46e9b8b077c46db8ba10,

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 constituent parsing and dependency parsing. The derivational model is accompanied by an equivalent probabilistic automaton model. 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 perplexity, as shown in experiments.",

keywords = "111 Mathematics, derivaatio, grammars, perplexity, 113 Computer and information sciences, parsing, grammars, weighted automata, perplexity, derivations, Probabilistic model, weighted automata, context-free derivation, discontinuity, derivation, 6121 Languages, dependenssikielioppi, j{\"a}sent{\"a}minen, sanaj{\"a}rjestys, kielen kompleksisuus, parsing, syntactic structures",

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

year = "2017",

month = mar,

day = "6",

doi = "10.1007/978-3-319-53733-7_22",

language = "English",

isbn = "978-3-319-53732-0",

series = "Lecture Notes in Computer Science",

publisher = "Springer-Verlag",

pages = "299--310",

editor = "Carlos Mart{\'i}n-Vide",

booktitle = "Language and Automata Theory and Applications",

address = "Germany",

}

TY - CHAP

T1 - A Derivational Model of Discontinuous Parsing

AU - Nederhof, Mark-Jan

AU - Yli-Jyrä, Anssi Mikael

PY - 2017/3/6

Y1 - 2017/3/6

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 constituent parsing and dependency parsing. The derivational model is accompanied by an equivalent probabilistic automaton model. 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 perplexity, as shown in experiments.

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 constituent parsing and dependency parsing. The derivational model is accompanied by an equivalent probabilistic automaton model. 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 perplexity, as shown in experiments.

KW - 111 Mathematics

KW - derivaatio

KW - grammars

KW - perplexity

KW - 113 Computer and information sciences

KW - parsing

KW - grammars

KW - weighted automata

KW - perplexity

KW - derivations

KW - Probabilistic model

KW - weighted automata

KW - context-free derivation

KW - discontinuity

KW - derivation

KW - 6121 Languages

KW - dependenssikielioppi

KW - jäsentäminen

KW - sanajärjestys

KW - kielen kompleksisuus

KW - parsing

KW - syntactic structures

U2 - 10.1007/978-3-319-53733-7_22

DO - 10.1007/978-3-319-53733-7_22

M3 - Chapter

SN - 978-3-319-53732-0

T3 - Lecture Notes in Computer Science

SP - 299

EP - 310

BT - Language and Automata Theory and Applications

A2 - Martín-Vide, Carlos

PB - Springer-Verlag

CY - Berlin

ER -

A Derivational Model of Discontinuous Parsing

Abstract

Publication series

Fields of Science

Access to Document

Other files and links

ADEQSYNTAX: A Usable Finite-State Model for Adequate Syntactic Complexity

Edward Gibson

Cite this

A Derivational Model of Discontinuous Parsing

Abstract

Publication series

Fields of Science

Access to Document

Other files and links

Projects

ADEQSYNTAX: A Usable Finite-State Model for Adequate Syntactic Complexity

Activities

Edward Gibson

Cite this