Estimating the Number of Equivalent Mutants

Abstract

Equivalent mutants are a constant source of aggravation in mutation testing because they distort mutation-based analysis; but the identification of equivalent mutants is known to be undecidable, in addition to being (in practice) tedious and error-prone. We argue that for most applications it is not necessary to individually identify equivalent mutants; rather it suffices to know/ estimate their number. In this paper, we discuss the specification and design of an automated tool that estimates the number of equivalent mutants generated from a base program by analyzing the source code of the program as well as the mutant generation policy.