期刊名称：International Journal of Soft Computing and Software Engineering
电子版ISSN：2251-7545
出版年度：2013
卷号：3
期号：3
页码：320-325
DOI：10.7321/jscse.v3.n3.48
出版社：Advance Academic Publisher
摘要：Graphical User Interface (GUI) is a mean of interaction between an end user and a software system. Software systems have gained an unprecedented popularity in last twenty years or so and the biggest factor behind this success is Graphical user interface. Software developing companies and teams have always shown a thirst for fully assured high quality software. To fulfill this deep desire of companies, software must go through an intensive testing, but it seems almost impossible to test a GUI application manually due to complexity involve in such effort. Obvious alternative is to go for automated testing. Models or Graphs are being considered as basis for automated GUI testing. Event-flow graph is one of several efforts towards automation of GUI testing. Thorough testing to satisfy the test organization or team’s demands is also a terminology, facing lack of consensus among different researchers. Usually test criterion corresponds a “coverage function” that measures how much of the automatically generated optimization parameters satisfies the given test criterion. Our past work has demonstrated that with the help of evolutionary algorithms and event flow representation we can get promising test coverage of GUI applications. Now we are going to extend our previous work and proposing the use of an evolutionary algorithm to gain multiple objectives. These objectives are to gain maximum coverage while keeping number of test cases at minimum side, and the evolutionary algorithm we are going to use for this purpose is Non-dominated Sorting Genetic Algorithm II ( NSGA-II).
