Automated Test Data Generation using Swarm Intelligence Approaches

摘要

Software engineers are facing uphill task of stabilizing software testing cost with ever increasing software complexity. Software test data generation is tedious, most time consuming, complex and central activity of testing. But this is also the only task in testing where automation can be deployed. Besides being NP-hard oroblem, software testing also requires exact solution making it more demanding as compared to other optimization problems. With this background, researchers are enthusiastically seeking employment of heuristic methods towards test data generation. This paper compares and evaluates two swarm intelligence based search techniques, namely, particle swarm optimization and artificial bee colony algorithm for automatic test data generation for procedure oriented programs using structural symbolic testing method. Test data is generated for each feasible path of the programs. Experiments on ten benchmark programs of varying sizes and complexities are conducted and the subsequent performance results are presented. The results of these approaches are also compared with genetic algorithm based technique for test data generation for demonstrating the efficiency of the swarm intelligence algorithms. The three algorithms have been evaluated on average test cases per path and average percentage coverage per path, it has been observed that the particle swarm optimization based algorithm outperforms the other two algorithms.