Continuous Rationale Identification in Issue Tracking and Version Control Systems

Anja Kleebaum, Barbara Paech, Jan Ole Johanssen, Bernd Bruegge

Research output: Contribution to journalConference articlepeer-review

Abstract

During the elicitation and implementation of requirements, the involved stakeholders make decisions and build up important decision knowledge, which they should make explicit and share. Issue tracking and version control systems offer opportunities for a lightweight capture of decision knowledge but currently lack techniques for making this knowledge explicit. In this paper, we present techniques to make decision knowledge explicit in the text of issue tracking and version control systems. We present features for its manual documentation as well as for its automatic identification using a text classifier. The text classifier identifies implicit decision knowledge in natural language texts, in particular, in the description and comments of tickets in the issue tracking system and commit messages.

Original languageEnglish
JournalCEUR Workshop Proceedings
Volume2857
StatePublished - 2021
Externally publishedYes
EventJoint Workshops of the 27th International Conference on Requirements Engineering, REFSQ 2021 - OpenRE, Posters and Tools Track, and Doctoral Symposium - Essen, Germany
Duration: 12 Apr 2021 → …

Keywords

  • ConDec
  • Decision Knowledge
  • Natural Language Processing
  • Rationale Management
  • Text Classification

Fingerprint

Dive into the research topics of 'Continuous Rationale Identification in Issue Tracking and Version Control Systems'. Together they form a unique fingerprint.

Cite this