Hedwig: A named entity linker

Marcus Klang; Pierre Nugues

Hedwig: A named entity linker

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

Abstract

Named entity linking is the task of identifying mentions of named things in text, such as “Barack Obama” or “New York”, and linking these mentions to unique identifiers. In this paper, we describe Hedwig, an end-to-end named entity linker, which uses a combination of word and character BILSTM models for mention detection, a Wikidata and Wikipedia-derived knowledge base with global information aggregated over nine language editions, and a PageRank algorithm for entity linking. We evaluated Hedwig on the TAC2017 dataset, consisting of news texts and discussion forums, and we obtained a final score of 59.9% on CEAFmC+, an improvement over our previous generation linker Ugglan, and a trilingual entity link score of 71.9%.

Original language	English
Title of host publication	LREC 2020 - 12th International Conference on Language Resources and Evaluation, Conference Proceedings
Editors	Nicoletta Calzolari, Frederic Bechet, Philippe Blache, Khalid Choukri, Christopher Cieri, Thierry Declerck, Sara Goggi, Hitoshi Isahara, Bente Maegaard, Joseph Mariani, Helene Mazo, Asuncion Moreno, Jan Odijk, Stelios Piperidis
Publisher	European Language Resources Association
Pages	4501-4508
Number of pages	8
ISBN (Electronic)	9791095546344
Publication status	Published - 2020
Event	12th International Conference on Language Resources and Evaluation, LREC 2020 - Marseille, France Duration: 2020 May 11 → 2020 May 16

Publication series

Name	LREC 2020 - 12th International Conference on Language Resources and Evaluation, Conference Proceedings

Conference

Conference	12th International Conference on Language Resources and Evaluation, LREC 2020
Country/Territory	France
City	Marseille
Period	2020/05/11 → 2020/05/16

Subject classification (UKÄ)

Computer Sciences
Natural Language Processing

Free keywords

Named entity annotation
Named entity linking
Named entity recognition

Access to Document

https://www.aclweb.org/anthology/2020.lrec-1.554.pdf

1 Doctoral Thesis (compilation)

Building Knowledge Graphs: Processing Infrastructure and Named Entity Linking
Klang, M., 2019 Sept 17, Ole Römers väg 3, Lund: Department of Computer Science, Lund University. 164 p.
Research output: Thesis › Doctoral Thesis (compilation)

Open Access
File

Cite this

Klang, M., & Nugues, P. (2020). Hedwig: A named entity linker. In N. Calzolari, F. Bechet, P. Blache, K. Choukri, C. Cieri, T. Declerck, S. Goggi, H. Isahara, B. Maegaard, J. Mariani, H. Mazo, A. Moreno, J. Odijk, & S. Piperidis (Eds.), LREC 2020 - 12th International Conference on Language Resources and Evaluation, Conference Proceedings (pp. 4501-4508). (LREC 2020 - 12th International Conference on Language Resources and Evaluation, Conference Proceedings). European Language Resources Association. https://www.aclweb.org/anthology/2020.lrec-1.554.pdf

Klang, Marcus ; Nugues, Pierre. / Hedwig : A named entity linker. LREC 2020 - 12th International Conference on Language Resources and Evaluation, Conference Proceedings. editor / Nicoletta Calzolari ; Frederic Bechet ; Philippe Blache ; Khalid Choukri ; Christopher Cieri ; Thierry Declerck ; Sara Goggi ; Hitoshi Isahara ; Bente Maegaard ; Joseph Mariani ; Helene Mazo ; Asuncion Moreno ; Jan Odijk ; Stelios Piperidis. European Language Resources Association, 2020. pp. 4501-4508 (LREC 2020 - 12th International Conference on Language Resources and Evaluation, Conference Proceedings).

@inproceedings{1bd6c3478ed24c32923fa352af1bff6a,

title = "Hedwig: A named entity linker",

abstract = "Named entity linking is the task of identifying mentions of named things in text, such as “Barack Obama” or “New York”, and linking these mentions to unique identifiers. In this paper, we describe Hedwig, an end-to-end named entity linker, which uses a combination of word and character BILSTM models for mention detection, a Wikidata and Wikipedia-derived knowledge base with global information aggregated over nine language editions, and a PageRank algorithm for entity linking. We evaluated Hedwig on the TAC2017 dataset, consisting of news texts and discussion forums, and we obtained a final score of 59.9% on CEAFmC+, an improvement over our previous generation linker Ugglan, and a trilingual entity link score of 71.9%.",

keywords = "Named entity annotation, Named entity linking, Named entity recognition",

author = "Marcus Klang and Pierre Nugues",

year = "2020",

language = "English",

series = "LREC 2020 - 12th International Conference on Language Resources and Evaluation, Conference Proceedings",

publisher = "European Language Resources Association",

pages = "4501--4508",

editor = "Nicoletta Calzolari and Frederic Bechet and Philippe Blache and Khalid Choukri and Christopher Cieri and Thierry Declerck and Sara Goggi and Hitoshi Isahara and Bente Maegaard and Joseph Mariani and Helene Mazo and Asuncion Moreno and Jan Odijk and Stelios Piperidis",

booktitle = "LREC 2020 - 12th International Conference on Language Resources and Evaluation, Conference Proceedings",

address = "France",

note = "12th International Conference on Language Resources and Evaluation, LREC 2020 ; Conference date: 11-05-2020 Through 16-05-2020",

}

Klang, M & Nugues, P 2020, Hedwig: A named entity linker. in N Calzolari, F Bechet, P Blache, K Choukri, C Cieri, T Declerck, S Goggi, H Isahara, B Maegaard, J Mariani, H Mazo, A Moreno, J Odijk & S Piperidis (eds), LREC 2020 - 12th International Conference on Language Resources and Evaluation, Conference Proceedings. LREC 2020 - 12th International Conference on Language Resources and Evaluation, Conference Proceedings, European Language Resources Association, pp. 4501-4508, 12th International Conference on Language Resources and Evaluation, LREC 2020, Marseille, France, 2020/05/11. <https://www.aclweb.org/anthology/2020.lrec-1.554.pdf>

Hedwig: A named entity linker. / Klang, Marcus ; Nugues, Pierre.
LREC 2020 - 12th International Conference on Language Resources and Evaluation, Conference Proceedings. ed. / Nicoletta Calzolari; Frederic Bechet; Philippe Blache; Khalid Choukri; Christopher Cieri; Thierry Declerck; Sara Goggi; Hitoshi Isahara; Bente Maegaard; Joseph Mariani; Helene Mazo; Asuncion Moreno; Jan Odijk; Stelios Piperidis. European Language Resources Association, 2020. p. 4501-4508 (LREC 2020 - 12th International Conference on Language Resources and Evaluation, Conference Proceedings).

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

TY - GEN

T1 - Hedwig

T2 - 12th International Conference on Language Resources and Evaluation, LREC 2020

AU - Klang, Marcus

AU - Nugues, Pierre

PY - 2020

Y1 - 2020

N2 - Named entity linking is the task of identifying mentions of named things in text, such as “Barack Obama” or “New York”, and linking these mentions to unique identifiers. In this paper, we describe Hedwig, an end-to-end named entity linker, which uses a combination of word and character BILSTM models for mention detection, a Wikidata and Wikipedia-derived knowledge base with global information aggregated over nine language editions, and a PageRank algorithm for entity linking. We evaluated Hedwig on the TAC2017 dataset, consisting of news texts and discussion forums, and we obtained a final score of 59.9% on CEAFmC+, an improvement over our previous generation linker Ugglan, and a trilingual entity link score of 71.9%.

AB - Named entity linking is the task of identifying mentions of named things in text, such as “Barack Obama” or “New York”, and linking these mentions to unique identifiers. In this paper, we describe Hedwig, an end-to-end named entity linker, which uses a combination of word and character BILSTM models for mention detection, a Wikidata and Wikipedia-derived knowledge base with global information aggregated over nine language editions, and a PageRank algorithm for entity linking. We evaluated Hedwig on the TAC2017 dataset, consisting of news texts and discussion forums, and we obtained a final score of 59.9% on CEAFmC+, an improvement over our previous generation linker Ugglan, and a trilingual entity link score of 71.9%.

KW - Named entity annotation

KW - Named entity linking

KW - Named entity recognition

M3 - Paper in conference proceeding

AN - SCOPUS:85096580434

T3 - LREC 2020 - 12th International Conference on Language Resources and Evaluation, Conference Proceedings

SP - 4501

EP - 4508

BT - LREC 2020 - 12th International Conference on Language Resources and Evaluation, Conference Proceedings

A2 - Calzolari, Nicoletta

A2 - Bechet, Frederic

A2 - Blache, Philippe

A2 - Choukri, Khalid

A2 - Cieri, Christopher

A2 - Declerck, Thierry

A2 - Goggi, Sara

A2 - Isahara, Hitoshi

A2 - Maegaard, Bente

A2 - Mariani, Joseph

A2 - Mazo, Helene

A2 - Moreno, Asuncion

A2 - Odijk, Jan

A2 - Piperidis, Stelios

PB - European Language Resources Association

Y2 - 11 May 2020 through 16 May 2020

ER -

Klang M , Nugues P. Hedwig: A named entity linker. In Calzolari N, Bechet F, Blache P, Choukri K, Cieri C, Declerck T, Goggi S, Isahara H, Maegaard B, Mariani J, Mazo H, Moreno A, Odijk J, Piperidis S, editors, LREC 2020 - 12th International Conference on Language Resources and Evaluation, Conference Proceedings. European Language Resources Association. 2020. p. 4501-4508. (LREC 2020 - 12th International Conference on Language Resources and Evaluation, Conference Proceedings).

Hedwig: A named entity linker

Abstract

Publication series

Conference

Subject classification (UKÄ)

Free keywords

Access to Document

Fingerprint

Research output

Building Knowledge Graphs: Processing Infrastructure and Named Entity Linking

Cite this