Authors : R. Uma Maheswari and D. Jeya Mala
Abstract: Changes in software can affect some of its behavior that had been implemented until that point. To discover such a problem, the perfect solution is testing the entire system completely again but there is a lack of time or resources for doing that. A different approach is to order the test cases and execute a subset of them which guarantees that the essential tests are executed first. Such an approach is known as regression test case prioritization (TCP). In this study, a new test suite prioritization technique which is a variant of Genetic Algorithm is presented. The proposed approach is based on Global Neighborhood Algorithm (GNA) which combines local search and global search. In the proposed system, the chromosomes represent the test case Sequence and gene represents the test case numbers. The aim is to find a test case sequence with 100% Fault Coverage in minimum time. Examination on whether the proposed approach overtakes existing test prioritization approach based on Genetic Algorithms (GA) for software test Prioritization is carried out. From the analysis of the results of the experiments, it is inferred that: computation time of GNA based Approach is very negligible over GA based Approach and GNA based approach is best in finding global optimal solution.
R. Uma Maheswari and D. Jeya Mala, 2016. GNA Based Test Case Prioritization. Asian Journal of Information Technology, 15: 1405-1410.