Ontology matching with Large Language Models and prioritized depth-first search

Ontology matching (OM) plays a key role in enabling data interoperability and knowledge sharing. Recently, methods based on Large Language Model (LLMs) have shown great promise in OM, particularly through the use of a retrieve-then-prompt pipeline. In this approach, relevant target entities are firs...

Descripción completa

Detalles Bibliográficos
Autores: Taboada Iglesias, María Jesús, Martínez Hernández, Diego, Arideh, Mohammed, Mosquera Losada, María Rosa
Tipo de recurso: artículo
Fecha de publicación:2025
País:España
Institución:Universidad de Santiago de Compostela (USC)
Repositorio:Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
Idioma:inglés
OAI Identifier:oai:minerva.usc.gal:10347/43542
Acceso en línea:https://hdl.handle.net/10347/43542
Access Level:acceso abierto
Palabra clave:Ontology matching
Retrieval augmented generation
Greedy search
Large Language Models
Zero-shot setting
id ES_2606fa2935eee439d543d33f19bb9a2d
oai_identifier_str oai:minerva.usc.gal:10347/43542
network_acronym_str ES
network_name_str España
repository_id_str
spelling Ontology matching with Large Language Models and prioritized depth-first searchTaboada Iglesias, María JesúsMartínez Hernández, DiegoArideh, MohammedMosquera Losada, María RosaOntology matchingRetrieval augmented generationGreedy searchLarge Language ModelsZero-shot settingOntology matching (OM) plays a key role in enabling data interoperability and knowledge sharing. Recently, methods based on Large Language Model (LLMs) have shown great promise in OM, particularly through the use of a retrieve-then-prompt pipeline. In this approach, relevant target entities are first retrieved and then used to prompt the LLM to predict the final matches. Despite their potential, these systems still present limited performance and high computational overhead. To address these issues, we introduce MILA, a novel approach that embeds a retrieve-identify-prompt pipeline within a prioritized depth-first search (PDFS) strategy. This approach efficiently identifies a large number of semantic correspondences with high accuracy, limiting LLM requests to only the most borderline cases. We evaluated MILA using three challenges from the 2024 edition of the Ontology Alignment Evaluation Initiative. Our method achieved the highest F-Measure in five of seven unsupervised tasks, outperforming state-of-the-art OM systems by up to 17%. It also performed better than or comparable to the leading supervised OM systems. MILA further exhibited task-agnostic performance, remaining stable across all tasks and settings, while significantly reducing runtime. These findings highlight that high-performance LLM-based OM can be achieved through a combination of programmed (PDFS), learned (embedding vectors), and prompting-based heuristics, without the need of domain-specific heuristics or fine-tuning.ElsevierUniversidade de Santiago de Compostela. Departamento de Electrónica e ComputaciónUniversidade de Santiago de Compostela. Departamento de Física AplicadaUniversidade de Santiago de Compostela. Departamento de Bioloxía Molecular20252025-05-0720252025-05-07journal articlehttp://purl.org/coar/resource_type/c_6501VoRhttp://purl.org/coar/version/c_970fb48d4fbd8a85info:eu-repo/semantics/articleapplication/pdfhttps://hdl.handle.net/10347/43542reponame:Minerva. Repositorio Institucional de la Universidad de Santiago de Compostelainstname:Universidad de Santiago de Compostela (USC)Inglésengopen accesshttp://purl.org/coar/access_right/c_abf2© 2025 The Authors. Published by Elsevier B.V. This is an open access article under the CC BY licensehttp://creativecommons.org/licenses/by/4.0/info:eu-repo/semantics/openAccessoai:minerva.usc.gal:10347/435422026-06-15T12:47:27Z
dc.title.none.fl_str_mv Ontology matching with Large Language Models and prioritized depth-first search
title Ontology matching with Large Language Models and prioritized depth-first search
spellingShingle Ontology matching with Large Language Models and prioritized depth-first search
Taboada Iglesias, María Jesús
Ontology matching
Retrieval augmented generation
Greedy search
Large Language Models
Zero-shot setting
title_short Ontology matching with Large Language Models and prioritized depth-first search
title_full Ontology matching with Large Language Models and prioritized depth-first search
title_fullStr Ontology matching with Large Language Models and prioritized depth-first search
title_full_unstemmed Ontology matching with Large Language Models and prioritized depth-first search
title_sort Ontology matching with Large Language Models and prioritized depth-first search
dc.creator.none.fl_str_mv Taboada Iglesias, María Jesús
Martínez Hernández, Diego
Arideh, Mohammed
Mosquera Losada, María Rosa
author Taboada Iglesias, María Jesús
author_facet Taboada Iglesias, María Jesús
Martínez Hernández, Diego
Arideh, Mohammed
Mosquera Losada, María Rosa
author_role author
author2 Martínez Hernández, Diego
Arideh, Mohammed
Mosquera Losada, María Rosa
author2_role author
author
author
dc.contributor.none.fl_str_mv Universidade de Santiago de Compostela. Departamento de Electrónica e Computación
Universidade de Santiago de Compostela. Departamento de Física Aplicada
Universidade de Santiago de Compostela. Departamento de Bioloxía Molecular
dc.subject.none.fl_str_mv Ontology matching
Retrieval augmented generation
Greedy search
Large Language Models
Zero-shot setting
topic Ontology matching
Retrieval augmented generation
Greedy search
Large Language Models
Zero-shot setting
description Ontology matching (OM) plays a key role in enabling data interoperability and knowledge sharing. Recently, methods based on Large Language Model (LLMs) have shown great promise in OM, particularly through the use of a retrieve-then-prompt pipeline. In this approach, relevant target entities are first retrieved and then used to prompt the LLM to predict the final matches. Despite their potential, these systems still present limited performance and high computational overhead. To address these issues, we introduce MILA, a novel approach that embeds a retrieve-identify-prompt pipeline within a prioritized depth-first search (PDFS) strategy. This approach efficiently identifies a large number of semantic correspondences with high accuracy, limiting LLM requests to only the most borderline cases. We evaluated MILA using three challenges from the 2024 edition of the Ontology Alignment Evaluation Initiative. Our method achieved the highest F-Measure in five of seven unsupervised tasks, outperforming state-of-the-art OM systems by up to 17%. It also performed better than or comparable to the leading supervised OM systems. MILA further exhibited task-agnostic performance, remaining stable across all tasks and settings, while significantly reducing runtime. These findings highlight that high-performance LLM-based OM can be achieved through a combination of programmed (PDFS), learned (embedding vectors), and prompting-based heuristics, without the need of domain-specific heuristics or fine-tuning.
publishDate 2025
dc.date.none.fl_str_mv 2025
2025-05-07
2025
2025-05-07
dc.type.none.fl_str_mv journal article
http://purl.org/coar/resource_type/c_6501
VoR
http://purl.org/coar/version/c_970fb48d4fbd8a85
dc.type.openaire.fl_str_mv info:eu-repo/semantics/article
format article
dc.identifier.none.fl_str_mv https://hdl.handle.net/10347/43542
url https://hdl.handle.net/10347/43542
dc.language.none.fl_str_mv Inglés
eng
language_invalid_str_mv Inglés
language eng
dc.rights.none.fl_str_mv open access
http://purl.org/coar/access_right/c_abf2
http://creativecommons.org/licenses/by/4.0/
dc.rights.openaire.fl_str_mv info:eu-repo/semantics/openAccess
rights_invalid_str_mv open access
http://purl.org/coar/access_right/c_abf2
http://creativecommons.org/licenses/by/4.0/
eu_rights_str_mv openAccess
dc.format.none.fl_str_mv application/pdf
dc.publisher.none.fl_str_mv Elsevier
publisher.none.fl_str_mv Elsevier
dc.source.none.fl_str_mv reponame:Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
instname:Universidad de Santiago de Compostela (USC)
instname_str Universidad de Santiago de Compostela (USC)
reponame_str Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
collection Minerva. Repositorio Institucional de la Universidad de Santiago de Compostela
repository.name.fl_str_mv
repository.mail.fl_str_mv
_version_ 1869404797490167808
score 15,81155

Similares en LA Referencia