Robust and Collective Entity Disambiguation through Semantic Embeddings

Stefan Zwicklbauer, Christin Seifert, Michael Granitzer

Research output: Chapter in Book/Report/Conference proceedingChapterAcademicpeer-review

38 Citations (Scopus)

Abstract

Entity disambiguation is the task of mapping ambiguous terms in natural-language text to its entities in a knowledge base. It finds its application in the extraction of structured data in RDF (Resource Description Framework) from textual documents, but equally so in facilitating artificial intelligence applications, such as Seman-tic Search, Reasoning and Question & Answering. We propose a new collective, graph-based disambiguation algorithm utilizing semantic entity and document embeddings for robust entity disam-biguation. Robust thereby refers to the property of achieving better than state-of-the-art results over a wide range of very different data sets. Our approach is also able to abstain if no appropriate entity can be found for a specific surface form. Our evaluation shows, that our approach achieves significantly (>5%) better results than all other publicly available disambiguation algorithms on 7 of 9 datasets without data set specific tuning. Moreover, we discuss the influence of the quality of the knowledge base on the disambigua-tion accuracy and indicate that our algorithm achieves better results than non-publicly available state-of-the-art algorithms.
Original languageEnglish
Title of host publicationSIGIR'16. Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval - SIGIR '16
Place of PublicationNew York
PublisherACM Press
Pages425-434
Number of pages10
ISBN (Print)9781450340694
DOIs
Publication statusPublished - 2016
Event39th International ACM SIGIR Conference on Research and Development in Information Retrieval 2016 - Palazzo dei Congressi di Pisa, Pisa, Italy
Duration: 17 Jul 201621 Jul 2016
Conference number: 39
http://sigir.org/sigir2016/programp/

Conference

Conference39th International ACM SIGIR Conference on Research and Development in Information Retrieval 2016
Abbreviated titleSIGIR 2016
CountryItaly
CityPisa
Period17/07/1621/07/16
Internet address

Fingerprint

Semantics
Artificial intelligence
Tuning

Cite this

Zwicklbauer, S., Seifert, C., & Granitzer, M. (2016). Robust and Collective Entity Disambiguation through Semantic Embeddings. In SIGIR'16. Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval - SIGIR '16 (pp. 425-434). New York: ACM Press. https://doi.org/10.1145/2911451.2911535
Zwicklbauer, Stefan ; Seifert, Christin ; Granitzer, Michael. / Robust and Collective Entity Disambiguation through Semantic Embeddings. SIGIR'16. Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval - SIGIR '16. New York : ACM Press, 2016. pp. 425-434
@inbook{89321c8ed027414c90a3ecc81fa7d775,
title = "Robust and Collective Entity Disambiguation through Semantic Embeddings",
abstract = "Entity disambiguation is the task of mapping ambiguous terms in natural-language text to its entities in a knowledge base. It finds its application in the extraction of structured data in RDF (Resource Description Framework) from textual documents, but equally so in facilitating artificial intelligence applications, such as Seman-tic Search, Reasoning and Question & Answering. We propose a new collective, graph-based disambiguation algorithm utilizing semantic entity and document embeddings for robust entity disam-biguation. Robust thereby refers to the property of achieving better than state-of-the-art results over a wide range of very different data sets. Our approach is also able to abstain if no appropriate entity can be found for a specific surface form. Our evaluation shows, that our approach achieves significantly (>5{\%}) better results than all other publicly available disambiguation algorithms on 7 of 9 datasets without data set specific tuning. Moreover, we discuss the influence of the quality of the knowledge base on the disambigua-tion accuracy and indicate that our algorithm achieves better results than non-publicly available state-of-the-art algorithms.",
author = "Stefan Zwicklbauer and Christin Seifert and Michael Granitzer",
year = "2016",
doi = "10.1145/2911451.2911535",
language = "English",
isbn = "9781450340694",
pages = "425--434",
booktitle = "SIGIR'16. Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval - SIGIR '16",
publisher = "ACM Press",

}

Zwicklbauer, S, Seifert, C & Granitzer, M 2016, Robust and Collective Entity Disambiguation through Semantic Embeddings. in SIGIR'16. Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval - SIGIR '16. ACM Press, New York, pp. 425-434, 39th International ACM SIGIR Conference on Research and Development in Information Retrieval 2016, Pisa, Italy, 17/07/16. https://doi.org/10.1145/2911451.2911535

Robust and Collective Entity Disambiguation through Semantic Embeddings. / Zwicklbauer, Stefan; Seifert, Christin; Granitzer, Michael.

SIGIR'16. Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval - SIGIR '16. New York : ACM Press, 2016. p. 425-434.

Research output: Chapter in Book/Report/Conference proceedingChapterAcademicpeer-review

TY - CHAP

T1 - Robust and Collective Entity Disambiguation through Semantic Embeddings

AU - Zwicklbauer, Stefan

AU - Seifert, Christin

AU - Granitzer, Michael

PY - 2016

Y1 - 2016

N2 - Entity disambiguation is the task of mapping ambiguous terms in natural-language text to its entities in a knowledge base. It finds its application in the extraction of structured data in RDF (Resource Description Framework) from textual documents, but equally so in facilitating artificial intelligence applications, such as Seman-tic Search, Reasoning and Question & Answering. We propose a new collective, graph-based disambiguation algorithm utilizing semantic entity and document embeddings for robust entity disam-biguation. Robust thereby refers to the property of achieving better than state-of-the-art results over a wide range of very different data sets. Our approach is also able to abstain if no appropriate entity can be found for a specific surface form. Our evaluation shows, that our approach achieves significantly (>5%) better results than all other publicly available disambiguation algorithms on 7 of 9 datasets without data set specific tuning. Moreover, we discuss the influence of the quality of the knowledge base on the disambigua-tion accuracy and indicate that our algorithm achieves better results than non-publicly available state-of-the-art algorithms.

AB - Entity disambiguation is the task of mapping ambiguous terms in natural-language text to its entities in a knowledge base. It finds its application in the extraction of structured data in RDF (Resource Description Framework) from textual documents, but equally so in facilitating artificial intelligence applications, such as Seman-tic Search, Reasoning and Question & Answering. We propose a new collective, graph-based disambiguation algorithm utilizing semantic entity and document embeddings for robust entity disam-biguation. Robust thereby refers to the property of achieving better than state-of-the-art results over a wide range of very different data sets. Our approach is also able to abstain if no appropriate entity can be found for a specific surface form. Our evaluation shows, that our approach achieves significantly (>5%) better results than all other publicly available disambiguation algorithms on 7 of 9 datasets without data set specific tuning. Moreover, we discuss the influence of the quality of the knowledge base on the disambigua-tion accuracy and indicate that our algorithm achieves better results than non-publicly available state-of-the-art algorithms.

U2 - 10.1145/2911451.2911535

DO - 10.1145/2911451.2911535

M3 - Chapter

SN - 9781450340694

SP - 425

EP - 434

BT - SIGIR'16. Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval - SIGIR '16

PB - ACM Press

CY - New York

ER -

Zwicklbauer S, Seifert C, Granitzer M. Robust and Collective Entity Disambiguation through Semantic Embeddings. In SIGIR'16. Proceedings of the 39th International ACM SIGIR conference on Research and Development in Information Retrieval - SIGIR '16. New York: ACM Press. 2016. p. 425-434 https://doi.org/10.1145/2911451.2911535