Automated optimum test case generation using web navigation graphs

Abstract

Increased diversity and complexity of software systems derived the need for test automation. Test Automation is the use of software for automatic execution of tests, comparison of results with expected outcome, setting up preconditions for test and test reporting functions. Model based testing is a test automation approach that generates and maintains more useful and flexible tests from explicit descriptions of the application. Graph theory techniques have been an important part of model based testing and several graph theory techniques have been proposed in the literature. We used a famous graph theory technique called maximum network flows for generating minimum number of test cases covering all features of a system. We used a web based case study to describe the working of the proposed optimum path finding algorithm. We found certain constraints on web navigation graph in order to completely reflect the system in the form of a graph. The resulting web navigation graph is given as input to the algorithm that we implemented, that returns the optimal test cases for the web application system. We then graphically showed the optimality and feature coverage of the algorithm with respect to the case study.