Poincaré Embeddings in the Task of Named Entity Recognition

David Muñoz; Fernando Pérez; David Pinto

doi:10.1007/978-3-030-60887-3_17

Poincaré Embeddings in the Task of Named Entity Recognition

David Muñoz, Fernando Pérez, David Pinto

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

Abstract

Hyperbolic embeddings have become important in many natural language processing tasks due to their great ability to capture latent hierarchical data and to encode valuable syntactic and semantic information. We study and consider the ability of Poincaré embeddings to get the most similar nodes to a given node when trying to recognize named entities in a set of text documents. In this paper, we propose a classifier model for the NER (Named Entity Recognition) task by implementing Poincaré embeddings and by using the most frequent n-grams and their Part-of-Speech (POS) structures from the training dataset. We found that POS structures and n-grams help to map possible named entities, while using Poincaré embeddings manage to affirm and refine this recognition, improving the recognition of named entities.

Original language	English
Title of host publication	Advances in Computational Intelligence - 19th Mexican International Conference on Artificial Intelligence, MICAI 2020, Proceedings
Editors	Lourdes Martínez-Villaseñor, Hiram Ponce, Oscar Herrera-Alcántara, Félix A. Castro-Espinoza
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	193-204
Number of pages	12
ISBN (Print)	9783030608866
DOIs	https://doi.org/10.1007/978-3-030-60887-3_17
Publication status	Published - 2020
Event	19th Mexican International Conference on Artificial Intelligence, MICAI 2020 - Mexico City, Mexico Duration: 12 Oct 2020 → 17 Oct 2020

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	12469 LNAI
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	19th Mexican International Conference on Artificial Intelligence, MICAI 2020
Country/Territory	Mexico
City	Mexico City
Period	12/10/20 → 17/10/20

Keywords

N-grams
NER
Named entity recognition
Part-of-Speech
Poincaré embeddings

Access to Document

10.1007/978-3-030-60887-3_17

https://arrow.tudublin.ie/ittscicon/23Licence: CC BY-NC-SA

Cite this

Muñoz, D., Pérez, F., & Pinto, D. (2020). Poincaré Embeddings in the Task of Named Entity Recognition. In L. Martínez-Villaseñor, H. Ponce, O. Herrera-Alcántara, & F. A. Castro-Espinoza (Eds.), Advances in Computational Intelligence - 19th Mexican International Conference on Artificial Intelligence, MICAI 2020, Proceedings (pp. 193-204). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12469 LNAI). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-60887-3_17

Muñoz, David ; Pérez, Fernando ; Pinto, David. / Poincaré Embeddings in the Task of Named Entity Recognition. Advances in Computational Intelligence - 19th Mexican International Conference on Artificial Intelligence, MICAI 2020, Proceedings. editor / Lourdes Martínez-Villaseñor ; Hiram Ponce ; Oscar Herrera-Alcántara ; Félix A. Castro-Espinoza. Springer Science and Business Media Deutschland GmbH, 2020. pp. 193-204 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{47c2e77dd21f4224b9e56e62f6c2a15c,

title = "Poincar{\'e} Embeddings in the Task of Named Entity Recognition",

abstract = "Hyperbolic embeddings have become important in many natural language processing tasks due to their great ability to capture latent hierarchical data and to encode valuable syntactic and semantic information. We study and consider the ability of Poincar{\'e} embeddings to get the most similar nodes to a given node when trying to recognize named entities in a set of text documents. In this paper, we propose a classifier model for the NER (Named Entity Recognition) task by implementing Poincar{\'e} embeddings and by using the most frequent n-grams and their Part-of-Speech (POS) structures from the training dataset. We found that POS structures and n-grams help to map possible named entities, while using Poincar{\'e} embeddings manage to affirm and refine this recognition, improving the recognition of named entities.",

keywords = "N-grams, NER, Named entity recognition, Part-of-Speech, Poincar{\'e} embeddings",

author = "David Mu{\~n}oz and Fernando P{\'e}rez and David Pinto",

note = "Publisher Copyright: {\textcopyright} 2020, Springer Nature Switzerland AG.; 19th Mexican International Conference on Artificial Intelligence, MICAI 2020 ; Conference date: 12-10-2020 Through 17-10-2020",

year = "2020",

doi = "10.1007/978-3-030-60887-3_17",

language = "English",

isbn = "9783030608866",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "193--204",

editor = "Lourdes Mart{\'i}nez-Villase{\~n}or and Hiram Ponce and Oscar Herrera-Alc{\'a}ntara and Castro-Espinoza, {F{\'e}lix A.}",

booktitle = "Advances in Computational Intelligence - 19th Mexican International Conference on Artificial Intelligence, MICAI 2020, Proceedings",

address = "Germany",

}

Muñoz, D, Pérez, F & Pinto, D 2020, Poincaré Embeddings in the Task of Named Entity Recognition. in L Martínez-Villaseñor, H Ponce, O Herrera-Alcántara & FA Castro-Espinoza (eds), Advances in Computational Intelligence - 19th Mexican International Conference on Artificial Intelligence, MICAI 2020, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12469 LNAI, Springer Science and Business Media Deutschland GmbH, pp. 193-204, 19th Mexican International Conference on Artificial Intelligence, MICAI 2020, Mexico City, Mexico, 12/10/20. https://doi.org/10.1007/978-3-030-60887-3_17

Poincaré Embeddings in the Task of Named Entity Recognition. / Muñoz, David; Pérez, Fernando; Pinto, David.
Advances in Computational Intelligence - 19th Mexican International Conference on Artificial Intelligence, MICAI 2020, Proceedings. ed. / Lourdes Martínez-Villaseñor; Hiram Ponce; Oscar Herrera-Alcántara; Félix A. Castro-Espinoza. Springer Science and Business Media Deutschland GmbH, 2020. p. 193-204 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12469 LNAI).

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

TY - GEN

T1 - Poincaré Embeddings in the Task of Named Entity Recognition

AU - Muñoz, David

AU - Pérez, Fernando

AU - Pinto, David

PY - 2020

Y1 - 2020

N2 - Hyperbolic embeddings have become important in many natural language processing tasks due to their great ability to capture latent hierarchical data and to encode valuable syntactic and semantic information. We study and consider the ability of Poincaré embeddings to get the most similar nodes to a given node when trying to recognize named entities in a set of text documents. In this paper, we propose a classifier model for the NER (Named Entity Recognition) task by implementing Poincaré embeddings and by using the most frequent n-grams and their Part-of-Speech (POS) structures from the training dataset. We found that POS structures and n-grams help to map possible named entities, while using Poincaré embeddings manage to affirm and refine this recognition, improving the recognition of named entities.

AB - Hyperbolic embeddings have become important in many natural language processing tasks due to their great ability to capture latent hierarchical data and to encode valuable syntactic and semantic information. We study and consider the ability of Poincaré embeddings to get the most similar nodes to a given node when trying to recognize named entities in a set of text documents. In this paper, we propose a classifier model for the NER (Named Entity Recognition) task by implementing Poincaré embeddings and by using the most frequent n-grams and their Part-of-Speech (POS) structures from the training dataset. We found that POS structures and n-grams help to map possible named entities, while using Poincaré embeddings manage to affirm and refine this recognition, improving the recognition of named entities.

KW - N-grams

KW - NER

KW - Named entity recognition

KW - Part-of-Speech

KW - Poincaré embeddings

UR - http://www.scopus.com/inward/record.url?scp=85092914239&partnerID=8YFLogxK

U2 - 10.1007/978-3-030-60887-3_17

DO - 10.1007/978-3-030-60887-3_17

M3 - Conference contribution

AN - SCOPUS:85092914239

SN - 9783030608866

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 193

EP - 204

BT - Advances in Computational Intelligence - 19th Mexican International Conference on Artificial Intelligence, MICAI 2020, Proceedings

A2 - Martínez-Villaseñor, Lourdes

A2 - Ponce, Hiram

A2 - Herrera-Alcántara, Oscar

A2 - Castro-Espinoza, Félix A.

PB - Springer Science and Business Media Deutschland GmbH

T2 - 19th Mexican International Conference on Artificial Intelligence, MICAI 2020

Y2 - 12 October 2020 through 17 October 2020

ER -

Muñoz D, Pérez F, Pinto D. Poincaré Embeddings in the Task of Named Entity Recognition. In Martínez-Villaseñor L, Ponce H, Herrera-Alcántara O, Castro-Espinoza FA, editors, Advances in Computational Intelligence - 19th Mexican International Conference on Artificial Intelligence, MICAI 2020, Proceedings. Springer Science and Business Media Deutschland GmbH. 2020. p. 193-204. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-60887-3_17

Poincaré Embeddings in the Task of Named Entity Recognition

Abstract

Publication series

Conference

Keywords

Access to Document

Fingerprint

Cite this