Exploring the use of large language models to build product Kansei semantic spaces
• GPT-4 can produce semantic spaces equivalent to those of existing techniques. • Two different approaches have been tested with similar positive results. • The extended procedure considers GPT-4 variability, allowing for more varied terms. • The short procedure is faster and produce terms slightly closer to the product.
Authors
- Jorge Alcaide-Marzal (ORCID: https://orcid.org/0000-0001-7116-9784)
- José Antonio Diego-Más (ORCID: https://orcid.org/0000-0002-3698-3411)
Institutions
- Universitat Politècnica de València (ES)
Publication Details
- Journal
- International Journal of Industrial Ergonomics
- Published
- 2025-03-17
- DOI
- https://doi.org/10.1016/j.ergon.2025.103709
- Citations
- 1
- Primary Topic
- Color perception and design
- Type
- article
- Field-Weighted Citation Impact
- 2.44