{"title":"An ontology-based retrieval augmented generation procedure for a voice-controlled maintenance assistant","authors":"Heiner Ludwig, Thorsten Schmidt, Mathias Kühn","doi":"10.1016/j.compind.2025.104289","DOIUrl":null,"url":null,"abstract":"<div><div>This paper presents a novel approach to support complex maintenance procedures through a dialogue-driven digital assistant using an ontology-based retrieval augmented generation method. The core of the proposed system relies on the strong formalisation capabilities of the graph-based Web Ontology Language (OWL), combined with various retrieval algorithms and different Large Language Models (LLMs) to determine the most useful context for answering user queries. To do this, we use the popular principle of Retrieval Augmented Generation (RAG). Graph traversal enriches the contextual knowledge, enabling more accurate and context-aware responses. An evaluation using an OWL example ontology and an extensive Q&A dataset demonstrates the improved retrieval quality achieved by combining classical and vector-based semantic matching methods. The community-driven analysis of generation quality illustrates the usability of an OWL-based assistant for maintenance procedures on the basis of contexts and LLMs of varying configurations.</div></div>","PeriodicalId":55219,"journal":{"name":"Computers in Industry","volume":"169 ","pages":"Article 104289"},"PeriodicalIF":8.2000,"publicationDate":"2025-04-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computers in Industry","FirstCategoryId":"94","ListUrlMain":"https://www.sciencedirect.com/science/article/pii/S0166361525000545","RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, INTERDISCIPLINARY APPLICATIONS","Score":null,"Total":0}
引用次数: 0
Abstract
This paper presents a novel approach to support complex maintenance procedures through a dialogue-driven digital assistant using an ontology-based retrieval augmented generation method. The core of the proposed system relies on the strong formalisation capabilities of the graph-based Web Ontology Language (OWL), combined with various retrieval algorithms and different Large Language Models (LLMs) to determine the most useful context for answering user queries. To do this, we use the popular principle of Retrieval Augmented Generation (RAG). Graph traversal enriches the contextual knowledge, enabling more accurate and context-aware responses. An evaluation using an OWL example ontology and an extensive Q&A dataset demonstrates the improved retrieval quality achieved by combining classical and vector-based semantic matching methods. The community-driven analysis of generation quality illustrates the usability of an OWL-based assistant for maintenance procedures on the basis of contexts and LLMs of varying configurations.
期刊介绍:
The objective of Computers in Industry is to present original, high-quality, application-oriented research papers that:
• Illuminate emerging trends and possibilities in the utilization of Information and Communication Technology in industry;
• Establish connections or integrations across various technology domains within the expansive realm of computer applications for industry;
• Foster connections or integrations across diverse application areas of ICT in industry.