Knowledge Tree
Ontology learning
Facts (13)
Sources
Construction of Knowledge Graphs: State and Challenges - arXiv arxiv.org 9 facts
perspectiveAl-Aswadi et al. argue that the field of ontology learning needs to transition from shallow learning to deep learning approaches to achieve deeper sentence analysis and improved learning of concepts and relations.
referenceMa and Molnár published 'Ontology learning from relational database: Opportunities for semantic information integration' in the Vietnam Journal of Computer Science in 2022.
claimOntology learning approaches can be categorized into linguistic approaches, which utilize NLP techniques like part-of-speech tagging and dependency analysis, and machine learning approaches.
referenceAl-Aswadi et al. provide a state-of-the-art overview of ontology learning from unstructured text, which aims to identify main concepts and relations for entities within a document collection.
referenceWong, Liu, and Bennamoun published 'Ontology learning from text: A look back and into the future' in ACM Computing Surveys in 2012.
referenceBrowarnik and Maimon published 'Ontology learning from text: why the ontology learning layer cake is not viable' in the International Journal of Signs and Semiotic Systems in 2015.
referenceMethods for learning ontologies from relational databases focus on reverse engineering or using mappings to transform a relational database schema into an ontology or knowledge graph. Reverse engineering allows for the derivation of an Entity-Relationship diagram or conceptual model from the relational schema, though this requires careful handling of trigger and constraint definitions to prevent semantic loss.
claimOntology learning consists of two main subfields: learning from unstructured text sources and learning from structured relational databases.
claimMachine learning approaches to ontology learning include statistic-based methods, such as co-occurrence analysis and clustering, and logic-based approaches, such as inductive logic programming or logical inference.
LLM-empowered knowledge graph construction: A survey - arXiv arxiv.org Oct 23, 2025 4 facts
referenceOntology learning approaches, which seek to derive ontological structures from textual corpora, were reviewed in Asim et al. (2018).
referenceNadeen Fathallah, Steffen Staab, and Alsayed Algergawy authored 'LLMs4Life: Large Language Models for Ontology Learning in Life Sciences', published as an arXiv preprint in December 2024.
referenceNadeen Fathallah, Arunav Das, Stefano De Giorgis, Andrea Poltronieri, Peter Haase, and Liubov Kovriguina authored 'NeOn-GPT: A Large Language Model-Powered Pipeline for Ontology Learning', published in the 2025 proceedings of The Semantic Web: ESWC 2024 Satellite Events.
referenceHamed Babaei Giglou, Jennifer D’Souza, and Sören Auer authored 'LLMs4OL: Large Language Models for Ontology Learning', published as an arXiv preprint in August 2023.