Short answer

Adopt a structured, prompt-driven approach when using LLMs for knowledge representation tasks in design, always incorporating expert review and validation.

Field
Classic Design
Source
Electronics (2025)
Method
Case study with generalized methodology development
Evidence
Strong effect

Large Language Models (LLMs) can significantly streamline the creation of ontologies by integrating with established engineering methodologies, provided a structured prompting approach and expert oversight are maintained. This classic design research insight is drawn from a 2025 study published in Electronics. Using Case study with generalized methodology development, researchers explored how this design variable affects real-world outcomes. The key design takeaway: Adopt a structured, prompt-driven approach when using LLMs for knowledge representation tasks in design, always incorporating expert review and validation.

Study
Classic DesignNew This WeekStrong effect

LLM-Assisted Ontology Engineering: A Structured Workflow for Knowledge Representation

Large Language Models (LLMs) can significantly streamline the creation of ontologies by integrating with established engineering methodologies, provided a structured prompting approach and expert oversight are maintained.

Electronics · 2025

01

Key Findings

  • 01A structured workflow integrating LLMs with classical ontology engineering steps is feasible and effective.
  • 02A guiding table for prompt engineering is crucial for consistency and efficiency when using LLMs for ontology generation.
  • 03Human domain expertise remains indispensable for ensuring conceptual rigor and practical relevance of the generated ontologies.
  • 04The generated ontology demonstrated logical consistency, structural properties, semantic accuracy, and inferential completeness.
02

Application

Design takeaway

Adopt a structured, prompt-driven approach when using LLMs for knowledge representation tasks in design, always incorporating expert review and validation.

How to apply

When developing a knowledge base for a new product feature or system, use an LLM to draft initial concepts and relationships, but meticulously structure your prompts and have domain experts review and refine every output.

Project actions

  • 01When defining your design problem, think about the knowledge you need to represent.
  • 02Consider how an LLM could help you brainstorm terms, relationships, or hierarchies for your design.
  • 03Document your prompts and the LLM's responses to show your process.
03

Method & Evidence

AimTo develop and validate a reusable methodology for ontology generation using Large Language Models (LLMs) guided by expert input and structured prompting.
MethodCase study with generalized methodology development
ProcedureA user preference ontology for adaptive interfaces in autonomous machines was developed using ChatGPT-4o. The process involved defining the domain, eliciting terms, constructing class hierarchies and properties, formalizing, populating, and validating the ontology. A guiding table was used to translate domain knowledge into structured prompts for the LLM, with continuous human expert involvement for refinement and validation.
ContextKnowledge representation, AI-assisted design, automotive sector (initial application)

Variables

IVStructured prompting strategy and LLM capabilities
DVOntology quality (logical consistency, structural properties, semantic accuracy, inferential completeness)
CVDomain definition, term elicitation process, human expert review process
04

Strengths & Limitations

Strengths

  • +Provides a concrete, replicable methodology for LLM-assisted ontology generation.
  • +Emphasizes the crucial role of human expertise in AI-driven knowledge engineering.

Limitations

LLMs can sometimes generate plausible but incorrect information, so rigorous checking by a human expert is always necessary. The quality of the output depends heavily on the quality of the input prompts.

Reliability & validity

The study validates the ontology using multiple metrics (logical consistency, structural properties, semantic accuracy, inferential completeness) and SPARQL queries, indicating strong internal validity. The generalizability of the methodology suggests good external validity for similar knowledge engineering tasks.

Think critically

To what extent can LLMs truly 'understand' domain-specific nuances, and where does the reliance on human expertise become a bottleneck rather than a safeguard?

05

Design Principles

"Leverage AI capabilities within a human-guided, structured workflow for robust knowledge engineering."

This research offers a practical framework for designers and engineers to leverage AI for complex knowledge representation tasks. By formalizing the process of ontology generation, it enables more efficient and consistent development of knowledge bases, crucial for intelligent systems, data integration, and design decision support.

06

What This Means for Your Design

This study shows how to use AI tools like ChatGPT to help build organized knowledge systems (ontologies) for design projects. It's like having a smart assistant, but you still need to guide it carefully with clear instructions and check its work.

How to use in your project

  • 1.Reference this study when explaining how you used AI tools to develop a knowledge base or conceptual model for your design project.
  • 2.Use the methodology's phases (domain definition, term elicitation, etc.) as a framework for your own knowledge-building process.
07

Add to My Project

08

Quick Cite

Paragraph starter

This research provides a valuable framework for leveraging Large Language Models (LLMs) in design knowledge engineering. The methodology, which integrates classical ontology engineering steps with LLM capabilities through structured prompting and expert validation, offers a systematic approach to building robust knowledge representations. This can be applied to design projects requiring the organization of complex information, such as user needs, material properties, or system functionalities, ensuring greater consistency and efficiency in knowledge management.

09

Source

Electronics

Methodological Exploration of Ontology Generation with a Dedicated Large Language Model

journal · 2025

View source

Questions About This Research

What does the research say about llm-assisted ontology engineering: a structured workflow for knowledge representation?
Adopt a structured, prompt-driven approach when using LLMs for knowledge representation tasks in design, always incorporating expert review and validation. Evidence: Electronics (2025).
Why does "LLM-Assisted Ontology Engineering: A Structured Workflow for Knowledge Representation" matter for design?
This research offers a practical framework for designers and engineers to leverage AI for complex knowledge representation tasks. By formalizing the process of ontology generation, it enables more efficient and consistent development of knowledge bases, crucial for intelligent systems, data integration, and design decision support.
How can designers apply this research?
Adopt a structured, prompt-driven approach when using LLMs for knowledge representation tasks in design, always incorporating expert review and validation.
What were the main findings?
A structured workflow integrating LLMs with classical ontology engineering steps is feasible and effective.. A guiding table for prompt engineering is crucial for consistency and efficiency when using LLMs for ontology generation.. Human domain expertise remains indispensable for ensuring conceptual rigor and practical relevance of the generated ontologies.. The generated ontology demonstrated logical consistency, structural properties, semantic accuracy, and inferential completeness.
What research method was used?
Case study with generalized methodology development.
How strong is the evidence?
Evidence strength is rated Strong effect, based on a 2025 journal from Electronics.
What should I do differently in my next project?
When developing a knowledge base for a new product feature or system, use an LLM to draft initial concepts and relationships, but meticulously structure your prompts and have domain experts review and refine every output.
What are the limitations?
The effectiveness of the methodology is dependent on the quality of the guiding table and the expertise of the human reviewers. Generalizability to highly abstract or rapidly evolving domains may require further adaptation.