Requirement based regression test suite reduction using dependence analysis

Title: Requirement based regression test suite reduction using dependence analysis
Authors: Xie, Bo
Date: 2006
Abstract: In this research, we present an extension to an existing requirement-based regression test suite reduction approach that uses EFSM model dependence analysis to reduce a given regression test suite [19]. The approach is based on the difference between the original model and the modified model expressed as a set of elementary model modifications: elementary addition of a transition and elementary deletion of a transition. For each elementary modification, the data and control dependencies are used to capture potential interactions between EFSM transitions. The potential interactions are used to reduce an existing regression test suite by eliminating repetitive tests. In this thesis, based on [19], we have defined some new dependencies introduced by elementary modifications of the EFSM model. We proposed algorithms to obtain these dependencies; to generate potential interactions with respect to (wrt) an elementary modification; and to reduce the regression test suite. We have also developed a Regression Test Suite Reduction tool, called RTSR based on these algorithms to be used for reducing the size of existing regression test suites. RTSR has been tested and validated. (Abstract shortened by UMI.)
CollectionTh├Ęses, 1910 - 2010 // Theses, 1910 - 2010
MR14974.PDF3.68 MBAdobe PDFOpen