Analyzing and Relating Bug Report Data for Feature Tracking

  • Authors:

  • Michael Fischer;Martin Pinzger;Harald Gall

  • Affiliations:

  • -;-;-

  • Venue:
  • WCRE '03 Proceedings of the 10th Working Conference on Reverse Engineering
  • Year:
  • 2003

Quantified Score

Hi-index 0.01

Visualization

Abstract

Gaining higher level evolutionary information aboutlarge software systems is a key in validating past and adjustingfuture development processes. In this paper, we analyzethe proximity of software features based on modificationand problem report data that capture the system's evolutionhistory. Features are instrumented and tracked, therelationships of modification and problem reports to thesefeatures are established, and the tracked features are visualizedto illustrate their otherwise hidden dependencies.Our approach uncovers these hidden relationships betweenfeatures via problem report analysis and presents them ineasy-to-evaluate visual form. Particular feature dependenciesthen can be selected to assess the feature evolution byzooming in into an arbitrary level of detail. Such visualizationof interwoven features, therefore, can indicate locationsof design erosion in the architectural evolution of asoftware system. Our approach has been validated usingthe large open source software project of Mozilla and itsbug reporting system Bugzilla.