Exploring Large Language Models for Ontology Alignment
Y. He, J. Chen, H. Dong, and I. Horrocks. (2023)cite arxiv:2309.07172Comment: Accepted at ISWC 2023 (Posters and Demos).
Abstract
This work investigates the applicability of recent generative Large Language
Models (LLMs), such as the GPT series and Flan-T5, to ontology alignment for
identifying concept equivalence mappings across ontologies. To test the
zero-shot performance of Flan-T5-XXL and GPT-3.5-turbo, we leverage challenging
subsets from two equivalence matching datasets of the OAEI Bio-ML track, taking
into account concept labels and structural contexts. Preliminary findings
suggest that LLMs have the potential to outperform existing ontology alignment
systems like BERTMap, given careful framework and prompt design.
Description
[2309.07172] Exploring Large Language Models for Ontology Alignment
%0 Generic
%1 he2023exploring
%A He, Yuan
%A Chen, Jiaoyan
%A Dong, Hang
%A Horrocks, Ian
%D 2023
%K biomedical llm myown oaei ontologies ontology-alignment ontology_mapping snomedct
%T Exploring Large Language Models for Ontology Alignment
%U http://arxiv.org/abs/2309.07172
%X This work investigates the applicability of recent generative Large Language
Models (LLMs), such as the GPT series and Flan-T5, to ontology alignment for
identifying concept equivalence mappings across ontologies. To test the
zero-shot performance of Flan-T5-XXL and GPT-3.5-turbo, we leverage challenging
subsets from two equivalence matching datasets of the OAEI Bio-ML track, taking
into account concept labels and structural contexts. Preliminary findings
suggest that LLMs have the potential to outperform existing ontology alignment
systems like BERTMap, given careful framework and prompt design.
@misc{he2023exploring,
abstract = {This work investigates the applicability of recent generative Large Language
Models (LLMs), such as the GPT series and Flan-T5, to ontology alignment for
identifying concept equivalence mappings across ontologies. To test the
zero-shot performance of Flan-T5-XXL and GPT-3.5-turbo, we leverage challenging
subsets from two equivalence matching datasets of the OAEI Bio-ML track, taking
into account concept labels and structural contexts. Preliminary findings
suggest that LLMs have the potential to outperform existing ontology alignment
systems like BERTMap, given careful framework and prompt design.},
added-at = {2023-09-22T12:57:44.000+0200},
author = {He, Yuan and Chen, Jiaoyan and Dong, Hang and Horrocks, Ian},
biburl = {https://www.bibsonomy.org/bibtex/212e3d21359e06e16093de53b7dbf3d92/hangdong},
description = {[2309.07172] Exploring Large Language Models for Ontology Alignment},
interhash = {5c0dff2be74f5098808466708dceba3a},
intrahash = {12e3d21359e06e16093de53b7dbf3d92},
keywords = {biomedical llm myown oaei ontologies ontology-alignment ontology_mapping snomedct},
note = {cite arxiv:2309.07172Comment: Accepted at ISWC 2023 (Posters and Demos)},
timestamp = {2023-09-22T12:57:44.000+0200},
title = {Exploring Large Language Models for Ontology Alignment},
url = {http://arxiv.org/abs/2309.07172},
year = 2023
}