Utilization of mutation testing data

Mutation testing is crucial for quality assurance but is often too slow and expensive due to the massive number of mutants generated. Existing methods like PMT and PMS offer some solutions, but a more aggressive approach to mutant reduction is urgently needed. This book introduces the MT-Forward Algorithm, a novel framework designed to dramatically reduce the cost and time of mutation testing. The method first clusters source code projects based on key metrics (SLOC, mutation score, etc.) for informed selection. The core of the MT-Forward Algorithm lies in its ability to streamline the testing process. It efficiently shrinks the scope from 21 initial mutation operators down to 7 priority operators. This translates to reducing the chosen mutant pool from 1019 live mutants to 749—a significant optimization. By leveraging features such as Coverage Test and LocMetrics Test, the resulting predictive model offers a major improvement over existing techniques. This book provides a practical, high-impact framework, enabling developers and QA teams to optimize their software testing lifecycle and deliver quality code faster and more affordably.