Abstract:
Graphical User Interfaces (GUIs) has become popular in today software products due to its user-friendly merit. However, GUI tests are costly and suffer explosive test cases. Moreover, software evolution causes the GUIs inevitably changed resulting in harder testing. Regression test has been widely used for GUIs software testing when software changes. It is considered to be the most expensive phase in the software testing process, though. Among several alternatives, the technique of test case prioritization would help identify the reduced set of test cases in the regression testing suite, still maintain the same criteria as the original number of test cases in order to solve the limitation of time and cost in this phase. This research thus proposes applying network centrality from social network analysis for ranking test case importance and finding suitable parameter(s) for GUI test case prioritization. The approach would lead to reduced resources during test execution and increased fault detection rate in early stages.
