基于Tent混沌的测试用例优先级排序

Test case prioritization based on Tent chaos

摘要:

针对标准粒子群算法（Particle Swarm Optimization，PSO）后期出现的早熟收敛，提出了一种基于Tent混沌的粒子群优化算法（Tent-Chaos Particle Swarm Optimization，TCPSO）用于测试用例优先级排序。首先，利用改进的Tent映射的三大特性初始化种群，使得粒子均匀分布，提高初始解的质量；并通过非线性递减的惯性权重函数对学习因子进行改进，以更新粒子速度与位置信息；其次，对陷入局部最优的粒子p_id进行混沌搜索，跳出局部最优，同时对当前种群中部分最差粒子p_iw进行混沌搜索，改善种群多样性；最后，采用测试用例的分支覆盖率和缺陷检测率作为评价标准，评判测试用例优劣程度。实验表明，提出的改进方法在分支覆盖率和缺陷检测率指标上均有优势。

Abstract:

Aiming at the premature convergence of the standard particle swarm optimization (PSO) algorithm, a new PSO algorithm based on Tent chaos(TCPSO) is proposed to prioritize test cases. Firstly, the population is initialized by using the randomness, ergodicity and regularity of the improved Tent map, so that the particles are evenly distributed and the quality of the initial solution is improved. At the same time, chaotic search is carried out for the optimal particle and some of the worst particle in the current population to improve the diversity of the population. Finally, the branch coverage of the test case and Defect detection rate are used as the evaluation criterion to judge the quality of the test case. Experiments show that the improved method has advantages in branch coverage and defect detection rate index.