Comparison of Mutation Testing and Manual Seeded Testing in Regression Test Selection

Abstract

Mutation testing and manual seeding testing are two distinct techniques employed in software testing. Mutation testing involves the introduction of artificial faults, called mutations, into the codebase, followed by executing the existing test suite against these mutated versions. On the other hand, manual seeding testing entails deliberately inserting faults into the source code by an individual or a group to monitor detection and removal rates and estimate the number of remaining faults. This paper introduces both techniques, discusses their advantages and disadvantages, and compares them in terms of test case selection using another testing technique, regression testing, which ensures that previously developed and tested software performs consistently after modifications and whose primary objective is to minimize testing costs by selecting and executing the minimum number of test cases.