用于测试用例生成的遗传算法改进

摘要

在软件测试技术中,高效的测试用例生成是简化测试工作、提高测试效率的必要手段.传统遗传算法是一种多点搜索和采用交叉操作的技术,有效应用于测试用例生成.遗传算法虽然具有良好的全局搜索能力,但对于局部空间的搜索问题不是很有效.针对此问题,对传统遗传算法在适应度函数、遗传算子方面进行改进,提高该算法在局部空间的搜索能力,提出一种基于改进遗传算法的测试用例生成方法.实验结果表明,所提出的方法在测试用例生成的效率和效果方面优于基于传统遗传算法的测试用例方法.%In the software testing technology, efficient test case generation is to simplify the testing, the necessary means to improve testing efficiency.Traditional genetic algorithm is a multi-point search and cross-operation technology, effectively used in the test case generation.Although the algorithm has good global search capability,but for the local search space problem is not very effective.In light of this, an improved genetic algorithm for testcase generation is presented, and improved on traditional genetic algorithm fitness function and genetic operators,to improve the algorithm's search abihty in the local space.Experimental results show that the algorithm efficiency and effectiveness of test case generation is better than traditional genetic algorithm.