基于Markov决策过程用交叉熵方法优化软件测试

被引:12
作者
张德平 [1 ,2 ]
聂长海 [1 ]
徐宝文 [1 ]
机构
[1] 东南大学计算机科学与工程学院
[2] 南京航空航天大学理学院
关键词
软件测试; 马尔可夫决策过程; 交叉熵方法; 最优测试剖面;
D O I
暂无
中图分类号
TP311.52 [];
学科分类号
081202 ; 0835 ;
摘要
研究了待测软件某些参数已知的条件下,以最小化平均测试费用为目标的软件测试优化问题.将软件测试过程处理成马尔可夫(Markov)决策过程,给出了软件测试的马尔可夫决策模型,运用交叉熵方法,通过一种学习策略获得软件测试的最优测试剖面,用于优化软件测试.模拟结果表明,学习策略给出的测试剖面要优于随机测试策略,检测和排除相同数目的软件缺陷,学习策略比随机测试能够显著地减少测试用例数,降低测试成本,提高缺陷检测效率.
引用
收藏
页码:2770 / 2779
页数:10
相关论文
共 11 条
[1]   划分测试中测试用例最优分配问题研究 [J].
张德平 ;
聂长海 ;
徐宝文 .
南京大学学报(自然科学版), 2005, (05) :553-561
[2]   The cross-entropy method for network reliability estimation [J].
Hui, KP ;
Bean, N ;
Kraetzl, M ;
Kroese, DP .
ANNALS OF OPERATIONS RESEARCH, 2005, 134 (01) :101-118
[3]   A tutorial on the cross-entropy method [J].
De Boer, PT ;
Kroese, DP ;
Mannor, S ;
Rubinstein, RY .
ANNALS OF OPERATIONS RESEARCH, 2005, 134 (01) :19-67
[4]   Importance sampling simulations of Markovian reliability systems using cross-entropy [J].
Ridder, A .
ANNALS OF OPERATIONS RESEARCH, 2005, 134 (01) :119-136
[5]   On the convergence of the cross-entropy method [J].
Margolin, L .
ANNALS OF OPERATIONS RESEARCH, 2005, 134 (01) :201-214
[6]   Optimal software testing and adaptive software testing in the context of software cybernetics [J].
Cai, KY .
INFORMATION AND SOFTWARE TECHNOLOGY, 2002, 44 (14) :841-855
[7]   An optimal statistical testing policy for software reliability demonstration of safety-critical systems [J].
Tal, O ;
McCollin, C ;
Bendell, A .
EUROPEAN JOURNAL OF OPERATIONAL RESEARCH, 2002, 137 (03) :544-557
[8]   Search-based software engineering [J].
Harman, M ;
Jones, BF .
INFORMATION AND SOFTWARE TECHNOLOGY, 2001, 43 (14) :833-839
[9]   Measuring complexity and coverage of software specifications [J].
Walton, G ;
Poore, JH .
INFORMATION AND SOFTWARE TECHNOLOGY, 2000, 42 (12) :859-872
[10]   The Cross-Entropy Method for Combinatorial and Continuous Optimization [J].
Reuven Rubinstein .
Methodology And Computing In Applied Probability, 1999, 1 (2) :127-190