Priority Path for Mutant Repairs on Mutation Testing
Sasa Ani Arnomo, Noraini Binti Ibrahim · 2019
Many operators are provided with the mutation testing tool. However, with many of the mutants tested it will produce many that are identified so that it requires a lot of time to repair the source code. The purpose of this study is to find the priority pathway for repairing mutants so that it is more effective when many mutants are detected on mutation testing. Mutation testing tools as material for evaluation of mutation operators are MUJAVA and PIT. The method used for evaluation is the dependency metric by including the evaluation of mutant similarities in the frequency and level of use of mutants. The results of the MUJAVA operator mutant test show that there is a priority path for mutant repair on mutation testing. The path starts from the first group improvement (AOIU, AOIS, ROR), the second group (COI, ASRS, JSD), and the third group (AORB, COR, LOI, JSI). Based on the number of followers of mutant operators, the efficiency level is 33%. The dependency value in PIT has a comprehensive value. The results show that the improved source code starts with the math mutator.