Towards Automated Ontology Generation from Unstructured Text: A Multi-Agent LLM Approach
Abid Talukder, Maruf Ahmed Mridul, Oshani Seneviratne
Why It Matters
What makes this one worth your time
Automating ontology generation can significantly streamline knowledge engineering processes, making it crucial for AI researchers and engineers focused on improving data organization and retrieval systems.
The paper proposes a multi-agent LLM architecture to enhance automated ontology generation from unstructured text.
Summary
The paper investigates automated ontology generation from unstructured text using a multi-agent large language model (LLM) approach. It identifies key failure modes in single-agent LLMs and proposes a multi-agent architecture with distinct roles to improve structural quality and queryability in ontology construction.
Key contributions
- Identification of key failure modes in single-agent LLMs for ontology generation.
- Introduction of a multi-agent architecture with distinct roles for ontology construction.
- Demonstration of improved structural quality and queryability using the proposed approach.
Notable insights
- Decomposing ontology construction into specific roles can improve structural quality.
- Front-loaded planning in multi-agent systems enhances the audibility and scalability of ontology generation.
Possible limitations
- Not stated in the abstract
Abstract
arXiv:2604.23090v1 Announce Type: new Abstract: Automatically generating formal ontologies from unstructured natural language remains a central challenge in knowledge engineering. While large language models (LLMs) show promise, it remains unclear which architectural design choices drive generation quality and why current approaches fail. We present a controlled experimental study using domain-specific insurance contracts to investigate these questions. We first establish a single-agent LLM baseline, identifying key failure modes such as poor Ontology Design Pattern compliance, structural redundancy, and ineffective iterative repair. We then introduce a multi-agent architecture that decomposes ontology construction into four artifact-driven roles: Domain Expert, Manager, Coder, and Quality Assurer. We evaluate performance across architectural quality (via a panel of heterogeneous LLM judges) and functional usability (via competency question driven SPARQL evaluation with complementary retrieval augmented generation based assessment). Results show that the multi-agent approach significantly improves structural quality and modestly enhances queryability, with gains driven primarily by front-loaded planning. These findings highlight planning-first, artifact-driven generation as a promising and more auditable path toward scalable automated ontology engineering.