Neighborhood mixture model for knowledge base completion

Dat Quoc Nguyen; Kairit Sirts; Lizhen Qu; Mark Johnson

doi:10.18653/v1/k16-1005

Neighborhood mixture model for knowledge base completion

Dat Quoc Nguyen, Kairit Sirts, Lizhen Qu, Mark Johnson

School of Computing

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

35 Citations (Scopus)

66 Downloads (Pure)

Abstract

Knowledge bases are useful resources for many natural language processing tasks, however, they are far from complete. In this paper, we define a novel entity representation as a mixture of its neighborhood in the knowledge base and apply this technique on TransE—a well-known embedding model for knowledge base completion. Experimental results show that the neighborhood information significantly helps to improve the results of the TransE, leading to better performance than obtained by other state-of-the-art embedding models on three benchmark datasets for triple classification, entity prediction and relation prediction tasks.

Original language	English
Title of host publication	Proceedings of the 20th SIGNLL Conference on Computational Natural Language Learning (CoNLL)
Place of Publication	Stroudsburg, PA
Publisher	Association for Computational Linguistics (ACL)
Pages	40-50
Number of pages	11
ISBN (Electronic)	9781945626197
DOIs	https://doi.org/10.18653/v1/k16-1005
Publication status	Published - 2016
Event	The SIGNLL Conference on Computational Natural Language Learning (20th : 2016) - Berlin, Germany Duration: 11 Aug 2016 → 12 Aug 2016 Conference number: 20th

Conference

Conference	The SIGNLL Conference on Computational Natural Language Learning (20th : 2016)
Abbreviated title	CoNLL 2016
Country/Territory	Germany
City	Berlin
Period	11/08/16 → 12/08/16

Bibliographical note

Copyright the Publisher 2016. 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.

Keywords

Knowledge base completion
embedding model
mixture model
link prediction
triple classification
entity prediction
relation prediction

Access to Document

10.18653/v1/k16-1005

Publisher versionFinal published version, 201 KBLicence: CC BY

Cite this

@inproceedings{384ce9d4ff704b2f8809e4df370ca557,

title = "Neighborhood mixture model for knowledge base completion",

abstract = "Knowledge bases are useful resources for many natural language processing tasks, however, they are far from complete. In this paper, we define a novel entity representation as a mixture of its neighborhood in the knowledge base and apply this technique on TransE—a well-known embedding model for knowledge base completion. Experimental results show that the neighborhood information significantly helps to improve the results of the TransE, leading to better performance than obtained by other state-of-the-art embedding models on three benchmark datasets for triple classification, entity prediction and relation prediction tasks.",

keywords = "Knowledge base completion, embedding model, mixture model, link prediction, triple classification, entity prediction, relation prediction",

author = "Nguyen, \{Dat Quoc\} and Kairit Sirts and Lizhen Qu and Mark Johnson",

note = "Copyright the Publisher 2016. 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.; The SIGNLL Conference on Computational Natural Language Learning (20th : 2016), CoNLL 2016 ; Conference date: 11-08-2016 Through 12-08-2016",

year = "2016",

doi = "10.18653/v1/k16-1005",

language = "English",

pages = "40--50",

booktitle = "Proceedings of the 20th SIGNLL Conference on Computational Natural Language Learning (CoNLL)",

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

}

Nguyen, DQ, Sirts, K, Qu, L & Johnson, M 2016, Neighborhood mixture model for knowledge base completion. in Proceedings of the 20th SIGNLL Conference on Computational Natural Language Learning (CoNLL). Association for Computational Linguistics (ACL), Stroudsburg, PA, pp. 40-50, The SIGNLL Conference on Computational Natural Language Learning (20th : 2016), Berlin, Germany, 11/08/16. https://doi.org/10.18653/v1/k16-1005

Neighborhood mixture model for knowledge base completion. / Nguyen, Dat Quoc; Sirts, Kairit; Qu, Lizhen et al.
Proceedings of the 20th SIGNLL Conference on Computational Natural Language Learning (CoNLL). Stroudsburg, PA: Association for Computational Linguistics (ACL), 2016. p. 40-50.

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

TY - GEN

T1 - Neighborhood mixture model for knowledge base completion

AU - Nguyen, Dat Quoc

AU - Sirts, Kairit

AU - Qu, Lizhen

AU - Johnson, Mark

N1 - Conference code: 20th

PY - 2016

Y1 - 2016

N2 - Knowledge bases are useful resources for many natural language processing tasks, however, they are far from complete. In this paper, we define a novel entity representation as a mixture of its neighborhood in the knowledge base and apply this technique on TransE—a well-known embedding model for knowledge base completion. Experimental results show that the neighborhood information significantly helps to improve the results of the TransE, leading to better performance than obtained by other state-of-the-art embedding models on three benchmark datasets for triple classification, entity prediction and relation prediction tasks.

AB - Knowledge bases are useful resources for many natural language processing tasks, however, they are far from complete. In this paper, we define a novel entity representation as a mixture of its neighborhood in the knowledge base and apply this technique on TransE—a well-known embedding model for knowledge base completion. Experimental results show that the neighborhood information significantly helps to improve the results of the TransE, leading to better performance than obtained by other state-of-the-art embedding models on three benchmark datasets for triple classification, entity prediction and relation prediction tasks.

KW - Knowledge base completion

KW - embedding model

KW - mixture model

KW - link prediction

KW - triple classification

KW - entity prediction

KW - relation prediction

UR - http://purl.org/au-research/grants/arc/DP160102156

UR - http://doi.org/10.18653/v1/K16-1005

UR - https://www.scopus.com/pages/publications/85018692437

U2 - 10.18653/v1/k16-1005

DO - 10.18653/v1/k16-1005

M3 - Conference proceeding contribution

SP - 40

EP - 50

BT - Proceedings of the 20th SIGNLL Conference on Computational Natural Language Learning (CoNLL)

PB - Association for Computational Linguistics (ACL)

CY - Stroudsburg, PA

T2 - The SIGNLL Conference on Computational Natural Language Learning (20th : 2016)

Y2 - 11 August 2016 through 12 August 2016

ER -

Neighborhood mixture model for knowledge base completion

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this