ConExion: Concept Extraction with Large Language Models

2025-04-17

AI Summary

This paper addresses the novel task of “universal concept extraction” for domain ontology construction—identifying *all* relevant concepts from documents (not merely salient phrases)—to support ontology coverage assessment and automated ontology learning. To this end, we propose the first unsupervised, large language model–based concept extraction framework, leveraging carefully engineered prompts to enable fine-grained, domain-adaptive concept discovery. Our approach requires neither annotated data nor domain-specific lexicons, ensuring strong generalizability and intrinsic interpretability. Evaluated on two established benchmarks, it achieves substantial F1-score improvements over prior state-of-the-art methods. To foster reproducibility and further research, we publicly release both the source code and benchmark datasets.

Technology Category

• Natural Language Processing: Information Extraction
• Knowledge Representation and Reasoning: Ontologies
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data

Abstract

In this paper, an approach for concept extraction from documents using pre-trained large language models (LLMs) is presented. Compared with conventional methods that extract keyphrases summarizing the important information discussed in a document, our approach tackles a more challenging task of extracting all present concepts related to the specific domain, not just the important ones. Through comprehensive evaluations of two widely used benchmark datasets, we demonstrate that our method improves the F1 score compared to state-of-the-art techniques. Additionally, we explore the potential of using prompts within these models for unsupervised concept extraction. The extracted concepts are intended to support domain coverage evaluation of ontologies and facilitate ontology learning, highlighting the effectiveness of LLMs in concept extraction tasks. Our source code and datasets are publicly available at https://github.com/ISE-FIZKarlsruhe/concept_extraction.

Problem

• Extracting all domain-related concepts from documents
• Improving F1 score over state-of-the-art techniques
• Exploring unsupervised concept extraction using LLM prompts

Innovation

• Uses pre-trained LLMs for concept extraction
• Extracts all domain concepts, not just keyphrases
• Explores prompts for unsupervised concept extraction