• English
  • Deutsch
  • Log In
    Password Login
    Research Outputs
    Fundings & Projects
    Researchers
    Institutes
    Statistics
Repository logo
Fraunhofer-Gesellschaft
  1. Home
  2. Fraunhofer-Gesellschaft
  3. Scopus
  4. Quest-RE QUestion Generation and Exploration STrategy for Requirements Engineering
 
  • Details
  • Full
Options
2024
Conference Paper
Title

Quest-RE QUestion Generation and Exploration STrategy for Requirements Engineering

Abstract
The quest for achieving completeness in requirements engineering (RE) is a complex challenge that requires innovative approaches to uncover and address hidden or incomplete requirements. This study introduces Quest- RE, a novel methodology leveraging Large Language Models (LLMs), specifically ChatGPT-4, to enhance the RE process through dynamic question generation and exploration strategies. By generating targeted questions referring to requirements, Quest- RE aims to improve communication between stakeholders and the RE team, thereby enhancing precision and completeness of requirements specifications (RS). The approach not only helps to identify gaps and missing elements in the requirements but also facilitates a deeper understanding and critical examination of the documented needs. For a given requirement, Quest- RE uses an algorithmic approach to generate related questions and question objectives. This ensures thorough exploration of each requirement beyond its initial scope. Two illustrative examples, one of it using requirements from the Fault Tolerant System Services (FTSS) and the Scheduling Services of NASA's X-38 Crew Return Vehicle, highlight the practical applicability and effectiveness of the approach in a complex engineering project. The study proposes that the incorporation of LLMs into RE can significantly influence communication and identification of requirements, thereby improving their overall quality. This research contributes to this area by offering an LLM - based approach to improve requirements elicitation and analysis, while also improving the readability and completeness of RS.
Author(s)
Hasso, Hussein
Fraunhofer-Institut für Kommunikation, Informationsverarbeitung und Ergonomie FKIE  
Fischer-Starcke, Bettina
Fraunhofer-Institut für Kommunikation, Informationsverarbeitung und Ergonomie FKIE  
Geppert, Hanna Claudia
Fraunhofer-Institut für Kommunikation, Informationsverarbeitung und Ergonomie FKIE  
Mainwork
32nd IEEE International Requirements Engineering Conference Workshops. Proceedings  
Conference
International Requirements Engineering Conference 2024  
DOI
10.1109/REW61692.2024.00006
Language
English
Fraunhofer-Institut für Kommunikation, Informationsverarbeitung und Ergonomie FKIE  
Keyword(s)
  • ChatGPT-4

  • Completeness of Requirements Specifications (RS)

  • Large Language Models (LLMs)

  • Question Generation

  • Read-ability of Requirements Specifications (RS)

  • Requirements Engineering (RE)

  • Cookie settings
  • Imprint
  • Privacy policy
  • Api
  • Contact
© 2024