改进教与学算法的两两组合测试数据生成

Improved pairwise test data generation of teaching-learning-based optimization algorithms

针对启发式算法生成两两组合测试数据集规模偏大、消耗时间过长的问题,提出一种改进教与学算法生成两两组合测试数据的方法。将教与学算法中的一个班级划分为多个班级,并使多个班级的教师之间进行相互学习,通过动态设定衰减阈值,使算法在教学完成后可以自适应的产生多条测试数据,同时结合one-test-at-a-time策略,缩短数据生成所消耗的时间。实验结果表明,该方法可以有效减少两两组合测试数据集的生成规模和消耗的时间。

The heuristic algorithm generates a large pair of test data sets and takes too long.To deal with this problem,an improved teaching and learning algorithm is proposed.In this algorithm,one class in the teaching and learning algorithm is divided into multiple classes,and the teachers in multiple classes are allowed to learn from each other.By dynamically setting the attenuation threshold,the algorithm can automatically generate multiple test data after the algorithm teaching is completed.At the same time,one-test-at-a-time strategy is combined to reduce the time consumed for data generation.Experimental results show that this method can effectively reduce the scale of data generation and the time consumed by the pairwise combinatorial test case suite.