基于决策树规则的回归测试技术研究  

Research of Regression Testing Technology Based on Rule of Decision-Tree

作　　者：廖敏[1] 李龙澍[1] 李森[1] 

机构地区：[1]安徽大学计算机系,安徽合肥230031

出　　处：《计算机技术与发展》2011年第5期25-28,共4页Computer Technology and Development

基　　金：安徽省自然科学基金(090412054)

摘　　要：回归测试中测试用例的优化选择是个关键环节,借助黑盒测试中的等价类划分选择测试用例可以提高测试的效率。文中介绍一种基于决策树规则的分类方法实现等价类的划分。该方法通过决策树提取规则,在按照一定的优先级对提取的决策树规则进行排序后,对测试用例库中的每个测试用例,选择优先级最高的规则进行匹配分类,最后从每一分类中选择具有代表性的测试用例,同时介绍了怎样构造该模型。该方法在保证了分类精度的同时能够提高测试的效率,该方法是有效的。The optimization choice of test case in regression testing is one of the key steps. Advance the efficiency of testing via equivalence partitioning method belonging to black box testing. There introduce a kind of classification method based on the decision tree algorithm to implement the sorting of equivalence class for advancing the efficient of testing. First get the decision rules via decision tree, and then choose the highest priority of rules to match the class for each test case in the test case library once or- dering the decision rules which were ordered according to the priority;At last choose the most representative test case. By the way, demonstrate how to build the model. The method in this article can improve the efficiency of testing at the moment of ensu- ring the accuracy of the classification and it is effective.

关 键 词：决策树 决策规则分类 回归测试 规则排序 测试用例 

分 类 号：TP31[自动化与计算机技术—计算机软件与理论]