Integrating lexical information into entity neighbourhood representations for relation prediction

Ian David Wood; Stephen Wan; Mark Johnson

doi:10.18653/v1/2021.naacl-main.268

Integrating lexical information into entity neighbourhood representations for relation prediction

Ian David Wood, Stephen Wan, Mark Johnson

School of Computing

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

3 Citations (Scopus)

96 Downloads (Pure)

Abstract

Relation prediction informed from a combination of text corpora and curated knowledge bases, combining knowledge graph completion with relation extraction, is a relatively little studied task. A system that can perform this task has the ability to extend an arbitrary set of relational database tables with information extracted from a document corpus. OpenKi[1] addresses this task through extraction of named entities and predicates via OpenIE tools then learning relation embeddings from the resulting entity-relation graph for relation prediction, outperforming previous approaches. We present an extension of OpenKi that incorporates embeddings of text-based representations of the entities and the relations. We demonstrate that this results in a substantial performance increase over a system without this information.

Original language	English
Title of host publication	The 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT 2021)
Subtitle of host publication	proceedings of the conference
Place of Publication	Stroudsburg, PA
Publisher	Association for Computational Linguistics (ACL)
Pages	3429-3436
Number of pages	8
ISBN (Electronic)	9781954085466
DOIs	https://doi.org/10.18653/v1/2021.naacl-main.268
Publication status	Published - 2021
Event	2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies - Online Duration: 6 Jun 2021 → 11 Jun 2021

Conference

Conference	2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
City	Online
Period	6/06/21 → 11/06/21

Bibliographical note

Copyright the Publisher 2021. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

Access to Document

10.18653/v1/2021.naacl-main.268

Publisher version (open access)Final published version, 319 KBLicence: CC BY

Cite this

Wood, I. D., Wan, S., & Johnson, M. (2021). Integrating lexical information into entity neighbourhood representations for relation prediction. In The 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT 2021): proceedings of the conference (pp. 3429-3436). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/2021.naacl-main.268

Wood, Ian David ; Wan, Stephen ; Johnson, Mark. / Integrating lexical information into entity neighbourhood representations for relation prediction. The 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT 2021): proceedings of the conference. Stroudsburg, PA : Association for Computational Linguistics (ACL), 2021. pp. 3429-3436

@inproceedings{2c60896e26744e358c601254bec148b9,

title = "Integrating lexical information into entity neighbourhood representations for relation prediction",

abstract = "Relation prediction informed from a combination of text corpora and curated knowledge bases, combining knowledge graph completion with relation extraction, is a relatively little studied task. A system that can perform this task has the ability to extend an arbitrary set of relational database tables with information extracted from a document corpus. OpenKi[1] addresses this task through extraction of named entities and predicates via OpenIE tools then learning relation embeddings from the resulting entity-relation graph for relation prediction, outperforming previous approaches. We present an extension of OpenKi that incorporates embeddings of text-based representations of the entities and the relations. We demonstrate that this results in a substantial performance increase over a system without this information.",

author = "Wood, {Ian David} and Stephen Wan and Mark Johnson",

note = "Copyright the Publisher 2021. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher. ; 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies ; Conference date: 06-06-2021 Through 11-06-2021",

year = "2021",

doi = "10.18653/v1/2021.naacl-main.268",

language = "English",

pages = "3429--3436",

booktitle = "The 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT 2021)",

publisher = "Association for Computational Linguistics (ACL)",

}

Wood, ID, Wan, S & Johnson, M 2021, Integrating lexical information into entity neighbourhood representations for relation prediction. in The 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT 2021): proceedings of the conference. Association for Computational Linguistics (ACL), Stroudsburg, PA, pp. 3429-3436, 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Online, 6/06/21. https://doi.org/10.18653/v1/2021.naacl-main.268

Integrating lexical information into entity neighbourhood representations for relation prediction. / Wood, Ian David; Wan, Stephen; Johnson, Mark.
The 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT 2021): proceedings of the conference. Stroudsburg, PA: Association for Computational Linguistics (ACL), 2021. p. 3429-3436.

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

TY - GEN

T1 - Integrating lexical information into entity neighbourhood representations for relation prediction

AU - Wood, Ian David

AU - Wan, Stephen

AU - Johnson, Mark

N1 - Copyright the Publisher 2021. Version archived for private and non-commercial use with the permission of the author/s and according to publisher conditions. For further rights please contact the publisher.

PY - 2021

Y1 - 2021

N2 - Relation prediction informed from a combination of text corpora and curated knowledge bases, combining knowledge graph completion with relation extraction, is a relatively little studied task. A system that can perform this task has the ability to extend an arbitrary set of relational database tables with information extracted from a document corpus. OpenKi[1] addresses this task through extraction of named entities and predicates via OpenIE tools then learning relation embeddings from the resulting entity-relation graph for relation prediction, outperforming previous approaches. We present an extension of OpenKi that incorporates embeddings of text-based representations of the entities and the relations. We demonstrate that this results in a substantial performance increase over a system without this information.

AB - Relation prediction informed from a combination of text corpora and curated knowledge bases, combining knowledge graph completion with relation extraction, is a relatively little studied task. A system that can perform this task has the ability to extend an arbitrary set of relational database tables with information extracted from a document corpus. OpenKi[1] addresses this task through extraction of named entities and predicates via OpenIE tools then learning relation embeddings from the resulting entity-relation graph for relation prediction, outperforming previous approaches. We present an extension of OpenKi that incorporates embeddings of text-based representations of the entities and the relations. We demonstrate that this results in a substantial performance increase over a system without this information.

UR - http://www.scopus.com/inward/record.url?scp=85137727793&partnerID=8YFLogxK

U2 - 10.18653/v1/2021.naacl-main.268

DO - 10.18653/v1/2021.naacl-main.268

M3 - Conference proceeding contribution

SP - 3429

EP - 3436

BT - The 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT 2021)

PB - Association for Computational Linguistics (ACL)

CY - Stroudsburg, PA

T2 - 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies

Y2 - 6 June 2021 through 11 June 2021

ER -

Wood ID, Wan S, Johnson M. Integrating lexical information into entity neighbourhood representations for relation prediction. In The 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL-HLT 2021): proceedings of the conference. Stroudsburg, PA: Association for Computational Linguistics (ACL). 2021. p. 3429-3436 doi: 10.18653/v1/2021.naacl-main.268

Integrating lexical information into entity neighbourhood representations for relation prediction

Abstract

Conference

Bibliographical note

Access to Document

Other files and links

Cite this