Optimization of Test Case Prioritization Using Automatic Dependency Detection

Abstract

Test case optimization can be viewed as a ‘multi-criteria decision-making’ problem in which one of the known benchmark is: dependencies; between test cases. In most of the events, with growth in size, interactions between components also increases, as they usually share information and sometimes even state too. Moreover, changes in terms of cost, schedule, requirements, and risk also affect interactions. Also, these dependences are inherited by test cases which intend to test functionality of a software. Dependent test cases have a huge and undeviating effect on the execution observations. It is not surprising, that dependencies if not managed properly give rise to numerous challenges to software testing. So, it has become necessary to realize interdependencies (intentional or unintentional) and assess their impact during regression testing to preserve software quality wise. Dependency structure-based information plays a vital part in boosting test case selection, minimization and prioritization during regression testing. Detecting dependencies manually need prior domain knowledge and can be time-consuming. Therefore, this paper suggest a methodology for test case prioritization using automatic dependency detection among test cases by examining coverage and requirement factors and then prioritizing test cases based on this information. For experimental analysis, ‘Eclipse Bug Defect Tracking’ data set is utilized and results show an increase in an enhanced ‘fault-detection rate’.