Automatic test data generation using genetic algorithm and program dependence graphs

@mastersthesis{citeulike:271389, abstract = {The complexity of software systems has been increasing dramatically in the past decade, and software testing as a most labor-intensive component becomes more and more expensive. Software costs at least 50\% of the total expense of software development, so any techniques leading to automatic generation of test data will have great potential to considerably reduce the cost. Existing approaches of automatic test data generation have achieved some success of using evolutionary computation algorithms as their optimization techniques to transform problems of test data generation to optimization problems, but they are unable to deal with Boolean variables or enumerated types and they need to be improved in many other aspects. This thesis presents a new approach utilizing program dependence analysis technique and genetic algorithms (GAs) to generate test data. A set of experiments using the new approach is reported to show its effectiveness and efficiency based on the established criterion.}, author = {Zhang, Hao }, citeulike-article-id = {271389}, keywords = {dissertation, testing}, posted-at = {2005-08-02 16:45:24}, priority = {2}, school = {UNIVERSITY OF ALBERTA (CANADA)}, title = {Automatic test data generation using genetic algorithm and program dependence graphs}, url = {http://wwwlib.umi.com/dissertations/fullcit/MQ96573}, year = {2004} }

TY - THES ID - citeulike:271389 L3 - citeulike-article-id:271389 TI - Automatic test data generation using genetic algorithm and program dependence graphs N2 - The complexity of software systems has been increasing dramatically in the past decade, and software testing as a most labor-intensive component becomes more and more expensive. Software costs at least 50% of the total expense of software development, so any techniques leading to automatic generation of test data will have great potential to considerably reduce the cost. Existing approaches of automatic test data generation have achieved some success of using evolutionary computation algorithms as their optimization techniques to transform problems of test data generation to optimization problems, but they are unable to deal with Boolean variables or enumerated types and they need to be improved in many other aspects. This thesis presents a new approach utilizing program dependence analysis technique and genetic algorithms (GAs) to generate test data. A set of experiments using the new approach is reported to show its effectiveness and efficiency based on the established criterion. KW - dissertation KW - testing AU - Zhang, Hao PY - 2004/// UR - http://wwwlib.umi.com/dissertations/fullcit/MQ96573 ER -