Automated Estimation of the Rate of Equivalent Mutants

Whereas most past research on mutation equivalence has focused on analyzing a base program and a mutant to determine whether they are equivalent, we favor an orthogonal approach: We argue that in addition to being difficult and error-prone, the determination of whether two programs are equivalent is also often unnecessary. For most practical applications, it is not necessary to identify equivalent mutants individually; it is sufficient to estimate their number. In this paper, we discuss an automated tool that does so by static analysis of the base program.