计算机工程
計算機工程
계산궤공정
COMPUTER ENGINEERING
2015年
4期
284-287,293
,共5页
组合测试%参数权值%参数需求度%测试用例%测试集%逐参数
組閤測試%參數權值%參數需求度%測試用例%測試集%逐參數
조합측시%삼수권치%삼수수구도%측시용례%측시집%축삼수
combinational test%parameter weight%parameter demand degree%test case%test set%In-Parameter-Order( IPO)
通常对组合测试研究的重点是生成最小的测试用例集,但其中却很少涉及到带权值的参数。针对带权值参数的两两组合测试用例生成问题,提出一种基于逐参数( IPO)策略的带权值参数两两组合测试用例生成算法。对影响IPO策略性能的3个影响因子进行改进,包括待扩展参数的扩展次序、已有测试集的扩展次序和待扩展参数的取值选择。在扩展完所有参数后,对此时的测试集使用约简算法进一步简化,得到按测试用例权值和降序排列的测试集。实验结果表明,该算法不仅能减少测试用例的生成数量,而且能解决参数的权值问题,使其在实际应用中可以更有效地降低测试成本。
通常對組閤測試研究的重點是生成最小的測試用例集,但其中卻很少涉及到帶權值的參數。針對帶權值參數的兩兩組閤測試用例生成問題,提齣一種基于逐參數( IPO)策略的帶權值參數兩兩組閤測試用例生成算法。對影響IPO策略性能的3箇影響因子進行改進,包括待擴展參數的擴展次序、已有測試集的擴展次序和待擴展參數的取值選擇。在擴展完所有參數後,對此時的測試集使用約簡算法進一步簡化,得到按測試用例權值和降序排列的測試集。實驗結果錶明,該算法不僅能減少測試用例的生成數量,而且能解決參數的權值問題,使其在實際應用中可以更有效地降低測試成本。
통상대조합측시연구적중점시생성최소적측시용례집,단기중각흔소섭급도대권치적삼수。침대대권치삼수적량량조합측시용례생성문제,제출일충기우축삼수( IPO)책략적대권치삼수량량조합측시용례생성산법。대영향IPO책략성능적3개영향인자진행개진,포괄대확전삼수적확전차서、이유측시집적확전차서화대확전삼수적취치선택。재확전완소유삼수후,대차시적측시집사용약간산법진일보간화,득도안측시용례권치화강서배렬적측시집。실험결과표명,해산법불부능감소측시용례적생성수량,이차능해결삼수적권치문제,사기재실제응용중가이경유효지강저측시성본。
The research of combinatorial test always focuses on how to generate a minimum set of test cases,but which rarely comes to weight values of parameters. This paper proposes an algorithm based on In-Parameter-Order ( IPO ) strategy to solve the problem of generating pairwises combinatorial test case set with weighted parameter. The algorithm makes some improvements on three factors which influence the performance of IPO strategy, including the order of expending of expanding parameter,the order of expending the existing test cases and the choose of the value of expanding parameter. In addition,after expending all parameters,it uses reduction algorithm processing the test set of this moment to simplify further. Finally,it gets the test set according to the descending order of weight values of test cases. Experimental result shows that this method not only can reduce the generation number of test cases,but also can solve the problem of weight values of parameters,which can more effectively reduce the cost of test in practical applications.