Gregory M. Kapfhammer

Associate Professor of Computer Science

http://www.cs.allegheny.edu/~gkapfham/

  
View  Edit  Uploads  History  Top  Changes  Search

Can Search-Based Prioritizers Improve the Coverage Effectiveness of Regression Test Suites?

Alexander P. Conrad and Gregory M. Kapfhammer. Can Search-Based Prioritizers Improve the Coverage Effectiveness of Regression Test Suites? Presented at the Centre for Research on Evolution, Search, and Testing, King's College London, London, England, United Kingdom, September 2008.

Abstract

This presentation describes test suite prioritization techniques that employ order-based genetic algorithms. We focus on evolutionary approaches that identify test orderings that quickly cover the test requirements and have high coverage effectiveness (CE) scores. After surveying the key data structures and operators used by the prioritizers, we give preliminary results concerning the efficiency and effectiveness of our search-based methods. As part of this empirical study, we use nine case study applications that contain up to 6,822 non-commented source statements (NCSS) and 110 test cases. The experimental results highlight the impact that the crossover operators (e.g., cycle and partially-mapped crossover), selection operators (e.g., roulette wheel and truncation), and stagnancy condition have on both the CE metric and execution time. For instance, we find that search-based prioritizers outperform random search methods and often find orderings that are equal to or better than those identified by greedy algorithms. Finally, the presentation concludes with a discussion of future research involving new genetic operators, case study applications, and statistical analysis techniques.

[PDF]

Links to this Page