COMPOSITE MUTANT: AN INNOVATIVE APPROACH TO MUTATION TESTING

摘要

This research concerns software testing using mutation analysis, a powerful technique for unit testing. The number of generated mutants can have a profound influence on the execution time. Our aim is to reduce the number of mutants without an effect on the effectiveness of testing. We propose an algorithm for generating composite mutants from independent mutants identified by program slicing technique. All of mutants are grouped with respect to slicing criterion into composite mutants. The experiments demonstrate that our method can save the number of mutants. Accordingly, the execution time to kill mutants can be reduced with this approach. We also prove that the number of test cases to kill composite mutants is not greater than the one to kill classic mutants.