Loading...

Automated Type and Priority Prediction of Issue Reports in Software Repositories

Akbari, Kiana | 2021

689 Viewed
  1. Type of Document: M.Sc. Thesis
  2. Language: Farsi
  3. Document No: 53769 (19)
  4. University: Sharif University of Technology
  5. Department: Computer Engineering
  6. Advisor(s): Heydarnoori, Abbas
  7. Abstract:
  8. Proper documentation plays an important role in successful software management and maintenance. Software repositories such as GitHub host an enormous number of software entities with various features. Developers collaboratively implement, use, and share these repositories in the community. Software repositories use issue tracking systems to keep track of issue reports, both to manage workload and document the highlight of teams’ effort. An issue report can contain a request for new features, a reported problem, or simply a question by users of a software product. As the number of these issues increases, it becomes harder to manage them. Github provides labels for tagging issues, however, only about 50% of issues have labels indicating the goal of an issue (feature request, bug report, support, and etc). In this work, we aim at automating the process of managing issue reports for software teams. We propose a two-stage approach to predict both the type and priority level an issue using state-of-the-art text classifiers and feature engineering methods. We train and evaluate our models in both intra- and inter-project settings. Our proposed approach can successfully predict type and priority of issue reports with 83% and 88% accuracy, respectively
  9. Keywords:
  10. Natural Language Processing ; Machine Learning ; Prioritization ; Classification ; Software Maintenance ; Safware Evolution ; Software Repositories ; Issue Reports

 Digital Object List

 Bookmark

...see more