SRDD (Software Requirement Description Dataset)

Welcome to SRDD (Software Requirement Description Dataset), a large prompted dataset tailored for Natural Language to Software (NL2Software) research. This repository contains a rich collection of prompts organized into 5 major categories and further subdivided into 40 subcategories. In total, the dataset comprises 1200 high-quality prompt samples extracted from ChatGPT 3.5, specifically curated to facilitate research in NL2Software.

Structure

• The generated prompt contains three parts:
  • Name of the software
  • Description of this software
  • Category of this software
• Details
  • check.csv # Check Results
    • The check csv file contains 14 columns, which are:
      • SoftwareCategory
      • SoftwareName
      • SoftwareDescription
      • Whether Obey Rule 1/2/3/4/5
      • Reason For Obeying(or not Obeying) Rule 1/2/3/4/5
      • Count of Rules Obeyed
    • The 5 rules are designed to make sure the generated software descriptions are clear to follow and easy to evaluate. Specifically, the 5 rules are:
      • Describe the primary function of this software, emphasizing its simplicity, commonality, and feasibility in implementation.
      • Craft a clear and comprehensive description that encapsulates all the essential information required to define the software's fundamental functionality.
      • Specify that the software does not require internet access, highlighting its self-contained nature.
      • This software can be realized without relying on real-world data sources.
      • Highlight the software's user-friendliness, emphasizing that it can be operated by a single individual and does not necessitate multiple users for testing, in contrast to online chat software.
  • data_ChatDev_format.sh # Data, in the format of executable shell scripts (in ChatDev)
  • data_attribute_format.csv # Data, in the format of csv, containing three columns, Name/Description/Category

Category

License

The SRDD dataset is licensed under CC BY-NC 4.0. This license explicitly permits non-commercial use of the data. We would like to emphasize that any models trained using these datasets should strictly adhere to the non-commercial usage restriction and should be employed exclusively for research purposes.