SEARCHING THE BEST TEST SUIT USING MUTATION TESTING AND MACHINE LEARNING APPROACH

The generation of test cases is a very time-consuming, crucial process optimized with different searching algorithms. Mutation testing helps to identify hidden defects not found using other test methods; this inspired the author to propose a hybrid algorithm known as the "Particle Swarm Shuffled Optimization Algorithm (PSSOA)" for generating test suits which are best for killing mutants. The proposed algorithm is devised by integrating the Particle Swarm and Shuffled Shepherd Algorithm and Adambased deep recurring neural networks (DRNN). In the optimization process, the proposed algorithm, namely PSSOA, is adapted for generating and finding the best test suite set that satisfies the tri-objectives like test suite size, mutant reduction rate and mutant score. The proposed hybrid approach is compared with the three most popular Genetic algorithm, Particle swarm and genetic algorithm, improved crow search algorithms to prove the improved performance and efficiency