Entity linking with a paraphrase flavor

Maria Pershina, Yifan He, Ralph Grishman

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The task of Named Entity Linking is to link entity mentions in the document to their correct entries in a knowledge base and to cluster NIL mentions. Ambiguous, misspelled, and incomplete entity mention names are the main challenges in the linking process. We propose a novel approach that combines two state-of-the-art models - for entity disambiguation and for paraphrase detection - to overcome these challenges. We consider name variations as paraphrases of the same entity mention and adopt a paraphrase model for this task. Our approach utilizes a graph-based disambiguation model based on Personalized Page Rank, and then refines and clusters its output using the paraphrase similarity between entity mention strings. It achieves a competitive performance of 80.5% in B3+F clustering score on diagnostic TAC EDL 2014 data.

Original languageEnglish (US)
Title of host publicationProceedings of the 10th International Conference on Language Resources and Evaluation, LREC 2016
PublisherEuropean Language Resources Association (ELRA)
Pages556-560
Number of pages5
ISBN (Electronic)9782951740891
StatePublished - Jan 1 2016
Event10th International Conference on Language Resources and Evaluation, LREC 2016 - Portoroz, Slovenia
Duration: May 23 2016May 28 2016

Other

Other10th International Conference on Language Resources and Evaluation, LREC 2016
CountrySlovenia
CityPortoroz
Period5/23/165/28/16

Fingerprint

diagnostic
performance
Paraphrase
Entity
Names
Disambiguation
Strings
Diagnostics
Graph
Incomplete

Keywords

  • Disambiguation
  • Linking
  • Pagerank

ASJC Scopus subject areas

  • Linguistics and Language
  • Library and Information Sciences
  • Language and Linguistics
  • Education

Cite this

Pershina, M., He, Y., & Grishman, R. (2016). Entity linking with a paraphrase flavor. In Proceedings of the 10th International Conference on Language Resources and Evaluation, LREC 2016 (pp. 556-560). European Language Resources Association (ELRA).

Entity linking with a paraphrase flavor. / Pershina, Maria; He, Yifan; Grishman, Ralph.

Proceedings of the 10th International Conference on Language Resources and Evaluation, LREC 2016. European Language Resources Association (ELRA), 2016. p. 556-560.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Pershina, M, He, Y & Grishman, R 2016, Entity linking with a paraphrase flavor. in Proceedings of the 10th International Conference on Language Resources and Evaluation, LREC 2016. European Language Resources Association (ELRA), pp. 556-560, 10th International Conference on Language Resources and Evaluation, LREC 2016, Portoroz, Slovenia, 5/23/16.
Pershina M, He Y, Grishman R. Entity linking with a paraphrase flavor. In Proceedings of the 10th International Conference on Language Resources and Evaluation, LREC 2016. European Language Resources Association (ELRA). 2016. p. 556-560
Pershina, Maria ; He, Yifan ; Grishman, Ralph. / Entity linking with a paraphrase flavor. Proceedings of the 10th International Conference on Language Resources and Evaluation, LREC 2016. European Language Resources Association (ELRA), 2016. pp. 556-560
@inproceedings{e2e29a3ff0be4b66845c7be3d0a168e5,
title = "Entity linking with a paraphrase flavor",
abstract = "The task of Named Entity Linking is to link entity mentions in the document to their correct entries in a knowledge base and to cluster NIL mentions. Ambiguous, misspelled, and incomplete entity mention names are the main challenges in the linking process. We propose a novel approach that combines two state-of-the-art models - for entity disambiguation and for paraphrase detection - to overcome these challenges. We consider name variations as paraphrases of the same entity mention and adopt a paraphrase model for this task. Our approach utilizes a graph-based disambiguation model based on Personalized Page Rank, and then refines and clusters its output using the paraphrase similarity between entity mention strings. It achieves a competitive performance of 80.5{\%} in B3+F clustering score on diagnostic TAC EDL 2014 data.",
keywords = "Disambiguation, Linking, Pagerank",
author = "Maria Pershina and Yifan He and Ralph Grishman",
year = "2016",
month = "1",
day = "1",
language = "English (US)",
pages = "556--560",
booktitle = "Proceedings of the 10th International Conference on Language Resources and Evaluation, LREC 2016",
publisher = "European Language Resources Association (ELRA)",

}

TY - GEN

T1 - Entity linking with a paraphrase flavor

AU - Pershina, Maria

AU - He, Yifan

AU - Grishman, Ralph

PY - 2016/1/1

Y1 - 2016/1/1

N2 - The task of Named Entity Linking is to link entity mentions in the document to their correct entries in a knowledge base and to cluster NIL mentions. Ambiguous, misspelled, and incomplete entity mention names are the main challenges in the linking process. We propose a novel approach that combines two state-of-the-art models - for entity disambiguation and for paraphrase detection - to overcome these challenges. We consider name variations as paraphrases of the same entity mention and adopt a paraphrase model for this task. Our approach utilizes a graph-based disambiguation model based on Personalized Page Rank, and then refines and clusters its output using the paraphrase similarity between entity mention strings. It achieves a competitive performance of 80.5% in B3+F clustering score on diagnostic TAC EDL 2014 data.

AB - The task of Named Entity Linking is to link entity mentions in the document to their correct entries in a knowledge base and to cluster NIL mentions. Ambiguous, misspelled, and incomplete entity mention names are the main challenges in the linking process. We propose a novel approach that combines two state-of-the-art models - for entity disambiguation and for paraphrase detection - to overcome these challenges. We consider name variations as paraphrases of the same entity mention and adopt a paraphrase model for this task. Our approach utilizes a graph-based disambiguation model based on Personalized Page Rank, and then refines and clusters its output using the paraphrase similarity between entity mention strings. It achieves a competitive performance of 80.5% in B3+F clustering score on diagnostic TAC EDL 2014 data.

KW - Disambiguation

KW - Linking

KW - Pagerank

UR - http://www.scopus.com/inward/record.url?scp=85037116397&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85037116397&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85037116397

SP - 556

EP - 560

BT - Proceedings of the 10th International Conference on Language Resources and Evaluation, LREC 2016

PB - European Language Resources Association (ELRA)

ER -