Entity Typing with Triples Using Language Models

verfasst von: Aniqa Riaz, Sara Abdollahi, Simon Gottschalk
Abstract: Entity Typing is the task of assigning a type to an entity in a knowledge graph. In this paper, we propose ETwT (Entity Typing with Triples), which leverages the triples of an entity, namely its label, description and the property labels used on it. We analyse which language models and classifiers are best suited to this input and compare ETwT’s performance on coarse-grained and fine-grained entity typing. Our evaluation demonstrates that ETwT is able to predict coarse-grained entity types with an F $$:1$$ score of 0.994, outperforming three baselines.
Organisationseinheit(en): Forschungszentrum L3S
Externe Organisation(en): Rheinische Friedrich-Wilhelms-Universität Bonn
Typ: Aufsatz in Konferenzband
Seiten: 169-173
Anzahl der Seiten: 5
Publikationsdatum: 21.10.2023
Publikationsstatus: Veröffentlicht
Peer-reviewed: Ja
ASJC Scopus Sachgebiete: Theoretische Informatik, Informatik (insg.)
Elektronische Version(en): https://doi.org/10.1007/978-3-031-43458-7_32 (Zugang: Geschlossen)

BibTeX

@inproceedings{4f3e7d524d2745839c74b0589d053817,
title = "Entity Typing with Triples Using Language Models",
abstract = "Entity Typing is the task of assigning a type to an entity in a knowledge graph. In this paper, we propose ETwT (Entity Typing with Triples), which leverages the triples of an entity, namely its label, description and the property labels used on it. We analyse which language models and classifiers are best suited to this input and compare ETwT{\textquoteright}s performance on coarse-grained and fine-grained entity typing. Our evaluation demonstrates that ETwT is able to predict coarse-grained entity types with an F $$:1$$ score of 0.994, outperforming three baselines.",
author = "Aniqa Riaz and Sara Abdollahi and Simon Gottschalk",
note = "Funding Information: This work was partially funded the Federal Ministry for Economic Affairs and Climate Action (BMWK), Germany (“ATTENTION!”, 01MJ22012D). ; 20th Extended Semantic Web Conference, ESWC 2023 ; Conference date: 28-05-2023 Through 01-06-2023",
year = "2023",
month = oct,
day = "21",
doi = "10.1007/978-3-031-43458-7_32",
language = "English",
isbn = "9783031434570",
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 = "169--173",
editor = "Catia Pesquita and Hala Skaf-Molli and Vasilis Efthymiou and Sabrina Kirrane and Axel Ngonga and Diego Collarana and Renato Cerqueira and Mehwish Alam and Cassia Trojahn and Sven Hertling",
booktitle = "The Semantic Web",
address = "Germany",
}