Automated Behavioral Regression Testing
AUTOMATED BEHAVIORAL REGRESSION TESTING - A Systematic & Comparative Study
Context : Automated Behavioral Regression testing(BERT) practices in industry have to be better understood &
implemented both for the industry itself and for the research community. When a program is modified during maintenance developers run the new version against its existing test suite to check that the changes made did not introduce unintended regression errors. The context is limited by the quality of existing test suite. Objective: It aims at surveying existing research on BERT testing in order to extend the importance of BERT and needs for future research. Method: A Systematic and comparative study is launched to find as much literature as possible and some of the research papers are classified with respect to focus, research type and contribution type. Results: The report generation by BERT can be improved widely by implementing various aggressive ways like abstracting, clustering and filtering and submitting the report to the developers which will helps the developers to identify regression errors with in less time and can proceed towards the solution as developer’s time is very critical in software development process. Conclusions: More techniques and evaluation research is needed to provide a better foundation for Behavioral Regression Testing.
Categories and Subject Descriptors: [Software Engineering]: Testing and Debugging,
Software Tools & Techniques
General Terms: Verification
Keywords: Regression testing, Software Evolution, Dynamic Analysis