spec-db

An open-source database of requirements specifications for research purposes

What should be included

• textual use-cases
• uml/emf models
• high-level requirements
• simple scripts
  • for generating simple reports
  • for running basic consistency checks

Directory structure

• Top-level directories represent individual projects
• Each project contains "original", "derived" and "generated" subdirs
  • original : contains original artefacts, such as PDF, ODT, DOC, HTML files ...
  • derived : contains artefacts derived from the original artefacts, usually by hand. These can be used for further machine processing
  • generated : artefact created using automated processing. These can be various reports, tables, graphs ...

Future work

• "build system" based on gradle or just a simple makefile
• more specifications (More POWER, Igor!)

Related Work

• https://github.com/rasbt/pattern_classification/
• http://mldata.org/