基于迭代预测降低巧合正确性测试用例影响的软件错误定位方法

A method based on iterative prediction to lower the effection from coincidental correctness

巧合正确性测试用例是指某个测试用例虽然在执行程序时覆盖了错误的代码行,但是其测试结果依然是正确的。在测试用例集中,巧合正确性测试用例是普遍存在的。巧合正确性测试用例对基于程序谱的软件错误定位方法的错误定位精度产生很大的影响。为了避免这一影响,本文提出一种基于迭代预测降低巧合正确性测试用例影响的方法。该方法的基本思想是通过迭代的方法,预测巧合正确性测试用例的数目N,再对候选测试用例的巧合正确性可疑值进行排序,去掉可疑值较高的前N个巧合正确性测试用例,利用新的测试集进行错误定位,直到找到错误语句,或者候选的巧合正确性测试用例的个数小于迭代预测值N为止。使用Siemens Suite测试用例集对系统进行了测试,测试结果表明该方法能够有效提高基于程序谱的软件错误定位方法的错误定位精度。

There are lots of coincidental correctness in tests suit. As a result of the existence of coincidental correctness, the softwareerror locating method based on the application spectrum of precision could be affected by a lot of coincidental correctness test cases.Therefore, this paper proposes a method based iterative prediction to improve the error code lines of dubious value. The main idea of themethod is by iterative method, prediction accuracy of coincidence N is the number of test cases, and then the correctness suspiciouscoincidence that test cases to the candidate value for sorting, get rid of dubious value higher N coincidence correctness before test cases,until you find false statements, or seeking candidates for the number of test cases is less than the number of iteration to predict N. Thesystem is tested by using Siemens Suite, and the results show that the method could effectively improve the software error locatingprecision based on the application spectrum.