论文:2012,Vol:30,Issue(2):296-300
引用本文:
郑炜, 杨威, 宋海静, 周南, 郭凯源. 一种基于规则提取的自动化测试用例生成方法[J]. 西北工业大学
Zheng Wei, Yang Wei, Song Haijing, Zhou Nan, Guo Kaiyuan. An Effective Algorithm for Automated Test Suite Generation Based on ANN(Artificial Neural Netwrok) Rule-Extraction[J]. Northwestern polytechnical university

一种基于规则提取的自动化测试用例生成方法
郑炜, 杨威, 宋海静, 周南, 郭凯源
西北工业大学 软件与微电子学院,陕西 西安 710072
摘要:
软件测试过程中,缩小测试用例集的规模,通常需要根据经验进行分析,找出输入输出间的非映射关系,这往往要耗费过多的测试资源。而人工神经网络在此问题的处理上有其特有的优势。因此文章提出了一种改进的规则提取方法,用于生成测试用例。通过构建神经网络模型,建立输入/输出之间的非线性映射关系,接着根据连接的权值,裁剪网络,去除与特定输出无关的输入属性。然后,在规则提取阶段仅保留两个与该输出最为相关的输入,并由此提取出IF-THEN规则,生成测试用例。文章完成了改进后规则提取算法各阶段的自动化,显著降低了在测试用例设计环节上的开销。最后,通过程序验证了该方法的有效性。
关键词:    自动化测试    测试用例生成    人工神经网络(ANN)    规则提取   
An Effective Algorithm for Automated Test Suite Generation Based on ANN(Artificial Neural Netwrok) Rule-Extraction
Zheng Wei, Yang Wei, Song Haijing, Zhou Nan, Guo Kaiyuan
Department of Software Engineering,Northwestern Polytechnical University,Xi'an 710072,China
Abstract:
Sections 1 and 2 of the full paper explain the algorithm mentioned in the title, which we believe is moreeffective than previous ones. Their core consists of:" Test suite minimization usually needs analyzing the nonlinearrelationships between input and output, which is an expensive, tedious and error-prone process in software testing.ANN has its unique advantages on these issues. We propose an approach that uses an improved ANN rule extractionmethod to generate test cases. For the specified output, this algortthm removes less significant inputs according totheir weight magnitudes. There will be only at most two inputs according to their specified output. Then, we extractIF-THEN rules from a pruned ANN, which can be used to generate test cases. And all these pruning phase and ruleextraction phase can be automated. " Section 3 uses a small program to prove its validity;after analysis, section 3presents the results in Table 2, thus showing that our algorithm reduces indeed considerably test domain analysis.
Key words:    algorithms    analysis    artificial intelligence    computer software    decision making    efficiency    evaluation    experiments    feature extraction    flowcharting    models    neural networks    nonlinear systems    sofewareengineering;ariticial neural network(ANN)    automated testing    rule-extraction    test suite generation   
收稿日期: 2011-05-19     修回日期:
DOI:
基金项目: 西北工业大学2011届本科毕业设计重点支持项目资助
通讯作者:     Email:
作者简介: 郑炜(1975-),西北工业大学副教授,主要从事软件测试、验证的研究。
相关功能
PDF(454KB) Free
打印本文
把本文推荐给朋友
作者相关文章
郑炜  在本刊中的所有文章
杨威  在本刊中的所有文章
宋海静  在本刊中的所有文章
周南  在本刊中的所有文章
郭凯源  在本刊中的所有文章

参考文献:
[1] Setiono R, Dejaeger K, et al. Software Effort Prediction Using Regression Rule Extraction from Neural Networks. InternationalConference on Tools with Artificial Intelligence, Proceedings, 2010, 22(2): 45-52
[2] Zhou Zhihua. Rule Extraction: Using Neural Networks or for Neural Networks? Journal of Computer Science and Technology, 2004, 19(2): 249-253
[3] Ulf Johansson, Tuve L, et al. Rule Extraction from Opaque Models-A Slightly Different Perspective. Proceedings of the 5 th In-ternational Conference on Machine Learning and Applications, 2006, 22-27
[4] David Martens, Bart Baesens, et al. Comprehensible Credit Scoring Models Using Rule Extraction from Support Vector Ma-chines. European Journal of Operational Research, 2007, 183(3): 1466 ~1476
[5] Vanmali M. Using a Neural Network in the Software Testing Process. 2002, University of South Florida