Digitala Vetenskapliga Arkivet

Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
From Natural Language to Domain Models: An Integration of ChatGPT into a Conceptual Modeling Tool
Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
2024 (English)Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
Abstract [en]

Conceptual modeling is by many considered an essential aspect of the software engineering practice, aiding in the documentation and design of complex systems. However, software engineers often abstain from modeling due to the short time frames of software projects and a lack of expertise regarding modeling. The emergence of Large Language models (LLMs) has introduced new possibilities for automating complex tasks, but the integration of LLMs into conceptual modeling tools is fairly unexplored. This thesis aims to bridge this gap by developing a modeling tool that uses ChatGPT and PlantUML, a tool that creates diagrams based on a domain-specific language, to automatically generate the Unified Modeling Language (UML) class diagrams based on natural language domain descriptions. Employing a design science framework, a document survey was conducted for problem explication, followed by semi-structured interviews for requirements elicitation. The tool was developed and evaluated through experiments with experienced modelers. The results showed that the tool sped up diagram creation, but the diagrams sometimes lacked accuracy. These inaccuracies suggest the tool is not suitable for inexperienced modelers in its current state, due to the false sense of security the generated diagrams might foster. Despite these limitations, the tool could still be useful to modelers by decreasing the workload involved in manually creating class diagrams. However, the limited sample size of ten subjects for the experiment raises questions surrounding the reliability and generalizability of the results. This suggests that further research is necessary to explore the tool’s potential in conceptual modeling.

Place, publisher, year, edition, pages
2024.
Keywords [en]
ChatGPT, Class diagrams, Unified Modeling Language, PlantUML, Artificial Intelligence, Conceptual modeling, Software engineering
National Category
Computer Sciences
Identifiers
URN: urn:nbn:se:su:diva-242725OAI: oai:DiVA.org:su-242725DiVA, id: diva2:1955657
Available from: 2025-04-30 Created: 2025-04-30

Open Access in DiVA

fulltext(4044 kB)16 downloads
File information
File name FULLTEXT01.pdfFile size 4044 kBChecksum SHA-512
d7bdd7c154492f7ca01987afb6c548a05d126fa362206fd92ff394127e27a5548e4ed066db1c90fc5e6a4a8d3efa11c73582733204a7075ed8ff6ae490744519
Type fulltextMimetype application/pdf

Search in DiVA

By author/editor
Lundell, StefanQvarnström, Minna
By organisation
Department of Computer and Systems Sciences
Computer Sciences

Search outside of DiVA

GoogleGoogle Scholar
Total: 16 downloads
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

urn-nbn

Altmetric score

urn-nbn
Total: 20 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf