Using History Information to Improve Design Flaws Detection

  • Authors:

  • Daniel Ratiu;Stéphane Ducasse;Tudor Gîrba;Radu Marinescu

  • Affiliations:

  • -;-;-;-

  • Venue:
  • CSMR '04 Proceedings of the Eighth Euromicro Working Conference on Software Maintenance and Reengineering (CSMR'04)
  • Year:
  • 2004

Quantified Score

Hi-index 0.00

Visualization

Abstract

As systems evolve and their structure decays, maintainersneed accurate and automatic identification of the designproblems. Current approaches for automatic detectionof design problems are not accurate enough because theyanalyze only a single version of a system and consequentlythey miss essential information as design problems appearand evolve over time. Our approach is to use the historicalinformation of the suspected flawed structure to increase theaccuracy of the automatic problem detection. Our means isto define measurements which summarize how persistent theproblem was and how much maintenance effort was spenton the suspected structure. We apply our approach on alarge scale case study and show how it improves the accuracyof the detection of God Classes and Data Classes, andadditionally how it adds valuable semantical informationabout the evolution of flawed design structures.