Resources and Evaluations for Danish Entity Resolution

Maria Jung Barrett; Hieu Lam; Martin Wu; Ophélie Lacroix; Barbara Plank; Anders Søgaard

Resources and Evaluations for Danish Entity Resolution

Maria Jung Barrett
, Hieu Lam
, Martin Wu
, Ophélie Lacroix
, Barbara Plank
, Anders Søgaard

NLPnorth

Alexandra Instituttet A/S
Københavns Universitet

Publikation: Konference artikel i Proceeding eller bog/rapport kapitel › Konferencebidrag i proceedings › Forskning › peer review

Abstract

Automatic coreference resolution is understudied in Danish even though most of the Danish Dependency Treebank (Buch-Kromann, 2003) is annotated with coreference relations. This paper describes a conversion of its partial, yet well-documented, coreference relations into coreference clusters and the training and evaluation of coreference models on this data. To the best of our knowledge, these are the first publicly available, neural coreference models for Danish. We also present a new entity linking annotation on the dataset using WikiData identifiers, a named entity disambiguation (NED) dataset, and a larger automatically created NED dataset enabling wikily supervised NED models. The entity linking annotation is benchmarked using a state-of-the-art neural entity disambiguation model.

Originalsprog	Engelsk
Titel	Fourth Workshop on Computational Models of Reference, Anaphora and Coreference (CRAC)
Forlag	Association for Computational Linguistics
Publikationsdato	2021
Sider	63–69
Status	Udgivet - 2021
Begivenhed	Computational Models of Reference, Anaphora and Coreference - Punta Cana, Dominica Varighed: 1 nov. 2021 → … Konferencens nummer: 4

Konference

Konference	Computational Models of Reference, Anaphora and Coreference
Nummer	4
Land/Område	Dominica
By	Punta Cana
Periode	01/11/2021 → …

Emneord

Coreference resolution
Danish language processing
Dependency Treebank
Neural models
Entity linking
WikiData
Named entity disambiguation (NED)
Wikily supervised models
Coreference clusters
Natural language processing (NLP)

Adgang til dokumentet

2021.crac-1.7Forlagets udgivne version, 231 KBLicens: CC BY

https://aclanthology.org/2021.crac-1.7.pdf

Citationsformater

@inproceedings{32b31ef46f3b44b1809b87bf07069577,

title = "Resources and Evaluations for Danish Entity Resolution",

abstract = "Automatic coreference resolution is understudied in Danish even though most of the Danish Dependency Treebank (Buch-Kromann, 2003) is annotated with coreference relations. This paper describes a conversion of its partial, yet well-documented, coreference relations into coreference clusters and the training and evaluation of coreference models on this data. To the best of our knowledge, these are the first publicly available, neural coreference models for Danish. We also present a new entity linking annotation on the dataset using WikiData identifiers, a named entity disambiguation (NED) dataset, and a larger automatically created NED dataset enabling wikily supervised NED models. The entity linking annotation is benchmarked using a state-of-the-art neural entity disambiguation model.",

keywords = "Coreference resolution, Danish language processing, Dependency Treebank, Neural models, Entity linking, WikiData, Named entity disambiguation (NED), Wikily supervised models, Coreference clusters, Natural language processing (NLP), Coreference resolution, Danish language processing, Dependency Treebank, Neural models, Entity linking, WikiData, Named entity disambiguation (NED), Wikily supervised models, Coreference clusters, Natural language processing (NLP)",

author = "Barrett, \{Maria Jung\} and Hieu Lam and Martin Wu and Oph{\'e}lie Lacroix and Barbara Plank and Anders S{\o}gaard",

year = "2021",

language = "English",

pages = "63–69",

booktitle = "Fourth Workshop on Computational Models of Reference, Anaphora and Coreference (CRAC)",

publisher = "Association for Computational Linguistics",

address = "United States",

note = "Computational Models of Reference, Anaphora and Coreference ; Conference date: 01-11-2021",

}

Barrett, MJ, Lam, H, Wu, M, Lacroix, O, Plank, B & Søgaard, A 2021, Resources and Evaluations for Danish Entity Resolution. i Fourth Workshop on Computational Models of Reference, Anaphora and Coreference (CRAC). Association for Computational Linguistics, s. 63–69, Computational Models of Reference, Anaphora and Coreference, Punta Cana, Dominica, 01/11/2021. <https://aclanthology.org/2021.crac-1.7.pdf>

Resources and Evaluations for Danish Entity Resolution. / Barrett, Maria Jung; Lam, Hieu; Wu, Martin et al.
Fourth Workshop on Computational Models of Reference, Anaphora and Coreference (CRAC). Association for Computational Linguistics, 2021. s. 63–69.

Publikation: Konference artikel i Proceeding eller bog/rapport kapitel › Konferencebidrag i proceedings › Forskning › peer review

TY - GEN

T1 - Resources and Evaluations for Danish Entity Resolution

AU - Barrett, Maria Jung

AU - Lam, Hieu

AU - Wu, Martin

AU - Lacroix, Ophélie

AU - Plank, Barbara

AU - Søgaard, Anders

N1 - Conference code: 4

PY - 2021

Y1 - 2021

N2 - Automatic coreference resolution is understudied in Danish even though most of the Danish Dependency Treebank (Buch-Kromann, 2003) is annotated with coreference relations. This paper describes a conversion of its partial, yet well-documented, coreference relations into coreference clusters and the training and evaluation of coreference models on this data. To the best of our knowledge, these are the first publicly available, neural coreference models for Danish. We also present a new entity linking annotation on the dataset using WikiData identifiers, a named entity disambiguation (NED) dataset, and a larger automatically created NED dataset enabling wikily supervised NED models. The entity linking annotation is benchmarked using a state-of-the-art neural entity disambiguation model.

AB - Automatic coreference resolution is understudied in Danish even though most of the Danish Dependency Treebank (Buch-Kromann, 2003) is annotated with coreference relations. This paper describes a conversion of its partial, yet well-documented, coreference relations into coreference clusters and the training and evaluation of coreference models on this data. To the best of our knowledge, these are the first publicly available, neural coreference models for Danish. We also present a new entity linking annotation on the dataset using WikiData identifiers, a named entity disambiguation (NED) dataset, and a larger automatically created NED dataset enabling wikily supervised NED models. The entity linking annotation is benchmarked using a state-of-the-art neural entity disambiguation model.

KW - Coreference resolution

KW - Danish language processing

KW - Dependency Treebank

KW - Neural models

KW - Entity linking

KW - WikiData

KW - Named entity disambiguation (NED)

KW - Wikily supervised models

KW - Coreference clusters

KW - Natural language processing (NLP)

KW - Coreference resolution

KW - Danish language processing

KW - Dependency Treebank

KW - Neural models

KW - Entity linking

KW - WikiData

KW - Named entity disambiguation (NED)

KW - Wikily supervised models

KW - Coreference clusters

KW - Natural language processing (NLP)

M3 - Article in proceedings

SP - 63

EP - 69

BT - Fourth Workshop on Computational Models of Reference, Anaphora and Coreference (CRAC)

PB - Association for Computational Linguistics

T2 - Computational Models of Reference, Anaphora and Coreference

Y2 - 1 November 2021

ER -