DBLPLink: An Entity Linker for the DBLP Scholarly Knowledge Graph

Debayan Banerjee; Arefa; Ricardo Usbeck; Chris Biemann

doi:10.48550/arXiv.2309.07545

DBLPLink: An Entity Linker for the DBLP Scholarly Knowledge Graph

Debayan Banerjee
, Arefa
, Ricardo Usbeck
, Chris Biemann

Research output: Contributions to collected editions/works › Article in conference proceedings › Research › peer-review

Abstract

In this work, we present a web application named DBLPLink, which performs entity linking over the DBLP scholarly knowledge graph. DBLPLink uses text-to-text pre-trained language models, such as T5, to produce entity label spans from an input text question. Entity candidates are fetched from a database based on the labels, and an entity re-ranker sorts them based on entity embeddings, such as TransE, DistMult and ComplEx. The results are displayed so that users may compare and contrast the results between T5-small, T5-base and the different KG embeddings used. The demo can be accessed at https://ltdemos.informatik.uni-hamburg.de/dblplink/. Code and data shall be made available at https://github.com/uhh-lt/dblplink.

Original language	English
Title of host publication	ISWC-Posters-Demos-Industry 2023 : Proceedings of the ISWC 2023 Posters, Demos and Industry Tracks: From Novel Ideas to Industrial Practice ISWC 2023
Editors	Irini Fundulaki, Kouji Kozaki, Daniel Garijo, Jose Manuel Gomez-Perez
Number of pages	5
Volume	3632
Publisher	Sun Site Central Europe (RWTH Aachen University)
Publication date	2023
DOIs	https://doi.org/10.48550/arXiv.2309.07545
Publication status	Published - 2023
Externally published	Yes
Event	22nd International Semantic Web Conference - ISWC 2023: From Novel Ideas to Industrial Practice - Athens, Greece Duration: 06.11.2023 → 10.11.2023 Conference number: 22 https://iswc2023.semanticweb.org/ https://doi.org/10.25798/ycnb-jn48 https://dblp.org/streams/conf/semweb#2023 http://www.wikidata.org/entity/Q119153957

Bibliographical note

Funding Information:
This research is performed as a part of the ARDIAS project, funded by the “Idea and Venture Fund“ research grant by Universität Hamburg, which is part of the Excellence Strategy of the Federal and State Governments. This work has additionally received funding through the German Research Foundation ?DFG) project NFDI4DS ?no. 460234259).

Publisher Copyright:
© 2023 Copyright for this paper by its authors.

Research areas and keywords

Informatics
Business informatics

Access to Document

10.48550/arXiv.2309.07545Licence: CC BY

Cite this

Banerjee, D., Arefa, Usbeck, R., & Biemann, C. (2023). DBLPLink: An Entity Linker for the DBLP Scholarly Knowledge Graph. In I. Fundulaki, K. Kozaki, D. Garijo, & J. M. Gomez-Perez (Eds.), ISWC-Posters-Demos-Industry 2023: Proceedings of the ISWC 2023 Posters, Demos and Industry Tracks: From Novel Ideas to Industrial Practice ISWC 2023 (Vol. 3632). (CEUR Workshop Proceedings; Vol. 3632). Sun Site Central Europe (RWTH Aachen University). https://doi.org/10.48550/arXiv.2309.07545

Banerjee, Debayan ; Arefa, ; Usbeck, Ricardo et al. / DBLPLink : An Entity Linker for the DBLP Scholarly Knowledge Graph. ISWC-Posters-Demos-Industry 2023: Proceedings of the ISWC 2023 Posters, Demos and Industry Tracks: From Novel Ideas to Industrial Practice ISWC 2023. editor / Irini Fundulaki ; Kouji Kozaki ; Daniel Garijo ; Jose Manuel Gomez-Perez. Vol. 3632 Sun Site Central Europe (RWTH Aachen University), 2023. (CEUR Workshop Proceedings).

@inbook{04e0c5b85fac41668ae11ede42b9bcea,

title = "DBLPLink: An Entity Linker for the DBLP Scholarly Knowledge Graph",

abstract = "In this work, we present a web application named DBLPLink, which performs entity linking over the DBLP scholarly knowledge graph. DBLPLink uses text-to-text pre-trained language models, such as T5, to produce entity label spans from an input text question. Entity candidates are fetched from a database based on the labels, and an entity re-ranker sorts them based on entity embeddings, such as TransE, DistMult and ComplEx. The results are displayed so that users may compare and contrast the results between T5-small, T5-base and the different KG embeddings used. The demo can be accessed at https://ltdemos.informatik.uni-hamburg.de/dblplink/. Code and data shall be made available at https://github.com/uhh-lt/dblplink.",

keywords = "Informatics, Business informatics",

author = "Debayan Banerjee and Arefa and Ricardo Usbeck and Chris Biemann",

note = "Funding Information: This research is performed as a part of the ARDIAS project, funded by the “Idea and Venture Fund“ research grant by Universit{\"a}t Hamburg, which is part of the Excellence Strategy of the Federal and State Governments. This work has additionally received funding through the German Research Foundation ?DFG) project NFDI4DS ?no. 460234259). Publisher Copyright: {\textcopyright} 2023 Copyright for this paper by its authors.; 22nd International Semantic Web Conference - ISWC 2023 : From Novel Ideas to Industrial Practice, ISWC 2023 ; Conference date: 06-11-2023 Through 10-11-2023",

year = "2023",

doi = "10.48550/arXiv.2309.07545",

language = "English",

volume = "3632",

series = "CEUR Workshop Proceedings",

publisher = "Sun Site Central Europe (RWTH Aachen University)",

editor = "Irini Fundulaki and Kouji Kozaki and Daniel Garijo and Gomez-Perez, \{Jose Manuel\}",

booktitle = "ISWC-Posters-Demos-Industry 2023",

address = "Germany",

url = "https://iswc2023.semanticweb.org/, https://doi.org/10.25798/ycnb-jn48, https://dblp.org/streams/conf/semweb\#2023, http://www.wikidata.org/entity/Q119153957",

}

Banerjee, D, Arefa, , Usbeck, R & Biemann, C 2023, DBLPLink: An Entity Linker for the DBLP Scholarly Knowledge Graph. in I Fundulaki, K Kozaki, D Garijo & JM Gomez-Perez (eds), ISWC-Posters-Demos-Industry 2023: Proceedings of the ISWC 2023 Posters, Demos and Industry Tracks: From Novel Ideas to Industrial Practice ISWC 2023. vol. 3632, CEUR Workshop Proceedings, vol. 3632, Sun Site Central Europe (RWTH Aachen University), 22nd International Semantic Web Conference - ISWC 2023, Athens, Greece, 06.11.23. https://doi.org/10.48550/arXiv.2309.07545

DBLPLink: An Entity Linker for the DBLP Scholarly Knowledge Graph. / Banerjee, Debayan; Arefa, ; Usbeck, Ricardo et al.
ISWC-Posters-Demos-Industry 2023: Proceedings of the ISWC 2023 Posters, Demos and Industry Tracks: From Novel Ideas to Industrial Practice ISWC 2023. ed. / Irini Fundulaki; Kouji Kozaki; Daniel Garijo; Jose Manuel Gomez-Perez. Vol. 3632 Sun Site Central Europe (RWTH Aachen University), 2023. (CEUR Workshop Proceedings; Vol. 3632).

Research output: Contributions to collected editions/works › Article in conference proceedings › Research › peer-review

TY - CHAP

T1 - DBLPLink

T2 - 22nd International Semantic Web Conference - ISWC 2023

AU - Banerjee, Debayan

AU - Arefa,

AU - Usbeck, Ricardo

AU - Biemann, Chris

N1 - Conference code: 22

PY - 2023

Y1 - 2023

N2 - In this work, we present a web application named DBLPLink, which performs entity linking over the DBLP scholarly knowledge graph. DBLPLink uses text-to-text pre-trained language models, such as T5, to produce entity label spans from an input text question. Entity candidates are fetched from a database based on the labels, and an entity re-ranker sorts them based on entity embeddings, such as TransE, DistMult and ComplEx. The results are displayed so that users may compare and contrast the results between T5-small, T5-base and the different KG embeddings used. The demo can be accessed at https://ltdemos.informatik.uni-hamburg.de/dblplink/. Code and data shall be made available at https://github.com/uhh-lt/dblplink.

AB - In this work, we present a web application named DBLPLink, which performs entity linking over the DBLP scholarly knowledge graph. DBLPLink uses text-to-text pre-trained language models, such as T5, to produce entity label spans from an input text question. Entity candidates are fetched from a database based on the labels, and an entity re-ranker sorts them based on entity embeddings, such as TransE, DistMult and ComplEx. The results are displayed so that users may compare and contrast the results between T5-small, T5-base and the different KG embeddings used. The demo can be accessed at https://ltdemos.informatik.uni-hamburg.de/dblplink/. Code and data shall be made available at https://github.com/uhh-lt/dblplink.

KW - Informatics

KW - Business informatics

UR - https://www.scopus.com/pages/publications/85184361288

U2 - 10.48550/arXiv.2309.07545

DO - 10.48550/arXiv.2309.07545

M3 - Article in conference proceedings

VL - 3632

T3 - CEUR Workshop Proceedings

BT - ISWC-Posters-Demos-Industry 2023

A2 - Fundulaki, Irini

A2 - Kozaki, Kouji

A2 - Garijo, Daniel

A2 - Gomez-Perez, Jose Manuel

PB - Sun Site Central Europe (RWTH Aachen University)

Y2 - 6 November 2023 through 10 November 2023

ER -

Banerjee D, Arefa , Usbeck R, Biemann C. DBLPLink: An Entity Linker for the DBLP Scholarly Knowledge Graph. In Fundulaki I, Kozaki K, Garijo D, Gomez-Perez JM, editors, ISWC-Posters-Demos-Industry 2023: Proceedings of the ISWC 2023 Posters, Demos and Industry Tracks: From Novel Ideas to Industrial Practice ISWC 2023. Vol. 3632. Sun Site Central Europe (RWTH Aachen University). 2023. (CEUR Workshop Proceedings). doi: 10.48550/arXiv.2309.07545

DBLPLink: An Entity Linker for the DBLP Scholarly Knowledge Graph

Abstract

Bibliographical note

Research areas and keywords

Access to Document

Fingerprint

Cite this