Automatic Software Prototype Data Generation for Rapid Requirements Validation

Strategic Insights -

Researchers from the Beijing University of Technology have proposed an automatic software prototype data generation method called InitialGPT, which aims to improve the efficiency and accuracy of requirements validation in software development. The method uses a prompt generation template, a data generation model, a data evaluation model, and multiple prototype data tools to automatically generate requirements-compliant prototype data. This approach has been validated on four real-world software system cases, showing significant improvements in efficiency and data quality.

Key Takeaways:

  • Researchers from the Beijing University of Technology proposed an automatic software prototype data generation method called InitialGPT to improve requirements validation efficiency and accuracy.
  • The method uses a prompt generation template, a data generation model, a data evaluation model, and multiple prototype data tools to automatically generate requirements-compliant prototype data.
  • InitialGPT has been validated on four real-world software system cases, showing a 7.02 times improvement in efficiency and generating data of similar quality to those written manually.
  • The approach has the potential to significantly improve the software development process by reducing manual effort and improving data accuracy.
  • The researchers involved in the study include Weiru Wang, Shuanglong Chang, and Juntao Gao from the Beijing University of Technology.
  • Keywords related to this research include Beijing, People's Republic of China, Asia, Electronics, Engineering, Software, and Beijing University of Technology.

Statistics:

  • The efficiency of requirements validation improved by a factor of 7.02 using the InitialGPT method.
  • The data generated by InitialGPT was found to be of similar quality to data written manually, but at a more advantageous cost and efficiency.
  • InitialGPT has been validated on four real-world software system cases.
  • The study was supported by the Guangdong Provincial Core Software Tackling Project of China and the Natural Science Foundation of Heilongjiang Province.
  • The method has demonstrated its potential for application in the computer software industry, with significant improvements in efficiency and data accuracy.

Sources:

  • VerticalNews (2025, October 14)
  • Electronics (2025;14(17))
  • Mdpi, St Alban-Anlage 66, Ch-4052 Basel, Switzerland
  • Weiru Wang, Beijing University of Technology, Faculty of Information Technology, School of Computing, Beijing 100124, People's Republic of China.