Neural Knowledge Base Repairs

Thomas Pellissier Tanon; Fabian Suchanek

doi:10.1007/978-3-030-77385-4_17

Neural Knowledge Base Repairs

Thomas Pellissier Tanon, Fabian Suchanek

Institut Polytechnique de Paris

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

Abstract

The curation of a knowledge base is a crucial but costly task. In this work, we suggest to make use of the advances in neural network research to improve the automated correction of constraint violations. Our method is a deep learning refinement of [23], and similarly uses the edits that solved some violations in the past to infer how to solve similar violations in the present. Our system makes use of the graph content, literal embeddings, and features extracted from Web pages to improve its performance. The experimental evaluation on Wikidata shows significant improvements over baselines.

Original language	English
Title of host publication	The Semantic Web - 18th International Conference, ESWC 2021, Proceedings
Editors	Ruben Verborgh, Katja Hose, Heiko Paulheim, Pierre-Antoine Champin, Maria Maleshkova, Oscar Corcho, Petar Ristoski, Mehwish Alam
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	287-303
Number of pages	17
ISBN (Print)	9783030773847
DOIs	https://doi.org/10.1007/978-3-030-77385-4_17
Publication status	Published - 1 Jan 2021
Event	18th European Semantic Web Conference, ESWC 2021 - Virtual, Online Duration: 6 Jun 2021 → 10 Jun 2021

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	12731 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	18th European Semantic Web Conference, ESWC 2021
City	Virtual, Online
Period	6/06/21 → 10/06/21

Access to Document

10.1007/978-3-030-77385-4_17

Cite this

Pellissier Tanon, T., & Suchanek, F. (2021). Neural Knowledge Base Repairs. In R. Verborgh, K. Hose, H. Paulheim, P.-A. Champin, M. Maleshkova, O. Corcho, P. Ristoski, & M. Alam (Eds.), The Semantic Web - 18th International Conference, ESWC 2021, Proceedings (pp. 287-303). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12731 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-030-77385-4_17

Pellissier Tanon, Thomas ; Suchanek, Fabian. / Neural Knowledge Base Repairs. The Semantic Web - 18th International Conference, ESWC 2021, Proceedings. editor / Ruben Verborgh ; Katja Hose ; Heiko Paulheim ; Pierre-Antoine Champin ; Maria Maleshkova ; Oscar Corcho ; Petar Ristoski ; Mehwish Alam. Springer Science and Business Media Deutschland GmbH, 2021. pp. 287-303 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{e6a8968b2a4f40d3aff69a1291868800,

title = "Neural Knowledge Base Repairs",

abstract = "The curation of a knowledge base is a crucial but costly task. In this work, we suggest to make use of the advances in neural network research to improve the automated correction of constraint violations. Our method is a deep learning refinement of [23], and similarly uses the edits that solved some violations in the past to infer how to solve similar violations in the present. Our system makes use of the graph content, literal embeddings, and features extracted from Web pages to improve its performance. The experimental evaluation on Wikidata shows significant improvements over baselines.",

author = "\{Pellissier Tanon\}, Thomas and Fabian Suchanek",

note = "Publisher Copyright: {\textcopyright} 2021, Springer Nature Switzerland AG.; 18th European Semantic Web Conference, ESWC 2021 ; Conference date: 06-06-2021 Through 10-06-2021",

year = "2021",

month = jan,

day = "1",

doi = "10.1007/978-3-030-77385-4\_17",

language = "English",

isbn = "9783030773847",

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 = "287--303",

editor = "Ruben Verborgh and Katja Hose and Heiko Paulheim and Pierre-Antoine Champin and Maria Maleshkova and Oscar Corcho and Petar Ristoski and Mehwish Alam",

booktitle = "The Semantic Web - 18th International Conference, ESWC 2021, Proceedings",

}

Pellissier Tanon, T & Suchanek, F 2021, Neural Knowledge Base Repairs. in R Verborgh, K Hose, H Paulheim, P-A Champin, M Maleshkova, O Corcho, P Ristoski & M Alam (eds), The Semantic Web - 18th International Conference, ESWC 2021, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 12731 LNCS, Springer Science and Business Media Deutschland GmbH, pp. 287-303, 18th European Semantic Web Conference, ESWC 2021, Virtual, Online, 6/06/21. https://doi.org/10.1007/978-3-030-77385-4_17

Neural Knowledge Base Repairs. / Pellissier Tanon, Thomas; Suchanek, Fabian.
The Semantic Web - 18th International Conference, ESWC 2021, Proceedings. ed. / Ruben Verborgh; Katja Hose; Heiko Paulheim; Pierre-Antoine Champin; Maria Maleshkova; Oscar Corcho; Petar Ristoski; Mehwish Alam. Springer Science and Business Media Deutschland GmbH, 2021. p. 287-303 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 12731 LNCS).

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

TY - GEN

T1 - Neural Knowledge Base Repairs

AU - Pellissier Tanon, Thomas

AU - Suchanek, Fabian

PY - 2021/1/1

Y1 - 2021/1/1

N2 - The curation of a knowledge base is a crucial but costly task. In this work, we suggest to make use of the advances in neural network research to improve the automated correction of constraint violations. Our method is a deep learning refinement of [23], and similarly uses the edits that solved some violations in the past to infer how to solve similar violations in the present. Our system makes use of the graph content, literal embeddings, and features extracted from Web pages to improve its performance. The experimental evaluation on Wikidata shows significant improvements over baselines.

AB - The curation of a knowledge base is a crucial but costly task. In this work, we suggest to make use of the advances in neural network research to improve the automated correction of constraint violations. Our method is a deep learning refinement of [23], and similarly uses the edits that solved some violations in the past to infer how to solve similar violations in the present. Our system makes use of the graph content, literal embeddings, and features extracted from Web pages to improve its performance. The experimental evaluation on Wikidata shows significant improvements over baselines.

U2 - 10.1007/978-3-030-77385-4_17

DO - 10.1007/978-3-030-77385-4_17

M3 - Conference contribution

AN - SCOPUS:85111077075

SN - 9783030773847

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

SP - 287

EP - 303

BT - The Semantic Web - 18th International Conference, ESWC 2021, Proceedings

A2 - Verborgh, Ruben

A2 - Hose, Katja

A2 - Paulheim, Heiko

A2 - Champin, Pierre-Antoine

A2 - Maleshkova, Maria

A2 - Corcho, Oscar

A2 - Ristoski, Petar

A2 - Alam, Mehwish

PB - Springer Science and Business Media Deutschland GmbH

T2 - 18th European Semantic Web Conference, ESWC 2021

Y2 - 6 June 2021 through 10 June 2021

ER -

Pellissier Tanon T, Suchanek F. Neural Knowledge Base Repairs. In Verborgh R, Hose K, Paulheim H, Champin PA, Maleshkova M, Corcho O, Ristoski P, Alam M, editors, The Semantic Web - 18th International Conference, ESWC 2021, Proceedings. Springer Science and Business Media Deutschland GmbH. 2021. p. 287-303. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-030-77385-4_17

Neural Knowledge Base Repairs

Abstract

Publication series

Conference

Access to Document

Fingerprint

Cite this