Genetic Algorithm and Its Application in Software Test Data Generation

Genetic Algorithm and Its Application in Software Test Data Generation

연구 분야: Verification

학회: 2023 International Conference on Applied Intelligence and Sustainable Computing (ICAISC)

Genetic algorithm is a computational technique that simulates the process of natural selection to solve complex problems. It has been widely applied in various fields, including software engineering. One of its applications in software engineering is test data generation. Test data generation is an important component of software testing, aimed at ensuring the quality and reliability of software systems. However, generating effective test data can be challenging and time-consuming, especially for complex systems with large input domains. Genetic algorithm provides a solution to this problem by using population based search methods to generate test data that meets specific criteria. This algorithm starts from the initial population of random test cases and iteratively evolves them through selection, crossover, and mutation operations until a satisfactory solution is found. The effectiveness of genetic algorithms in generating test data has been proven in various studies. It has been proven to generate various effective test cases, covering different parts of the input domain, and revealing hidden faults in the system. In short, genetic algorithms are a powerful tool for software engineers to generate effective test data for complex systems. Its application can significantly improve the quality and reliability of software systems by identifying hidden faults that may not have been discovered. In this study, we analyzed genetic algorithms in current software testing, controlled them based on model data and generation structure in software testing, and obtained effective calculation methods based on various genetic algorithms. By using traditional genetic algorithms and various swarm genetic algorithms to analyze the distribution and optimal value interval of two curves, it is shown that various swarm genetic methods have strong optimization ability, high accuracy, and can quickly jump out of local optima to obtain the final solution. They are a very effective optimizat... Show More