Abstract

Regression Testing is a performed to ensure modified code does not have any unintended side effect on the software. If regression testing is performed with retest-all method it will be very t ime consuming as testing activity. Therefore test suite reduction methods are used to reduce the size of original test suite. Object ive of test suite reduction is to reduce those test cases which are redundant or less important in their fau lt revealing capability. Test suite reduction can only be used when time is critical to run all test cases and selective testing can only be done. Various methods exist in the literature related to test suite reduction of traditional software. Most of the methods are based of single objective optimization. In case of mult i ob jective optimization o f test suite, usually researchers assign different weight values to different objectives and co mbine them as single object ive. However in test suite reduction mu ltiple Pareto-optimal solutions are present, it is difficult to select one test case over other. Since GUI based software is our concern there exist very few reduction techniques and none of them consider mult iple objective based reduction. In this work we propose a new test suite reduction technique based on two objectives, event weight and number of faults identified by test case. We evaluated our results for 2 different applications and we achieved 20% reduction in test suite size for both applications. In Terp Paint 3.0 application co mpro mise 15.6% fau lt revealing capability and for Notepad 11.1% fau lt revealing capability is reduced.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call