Automatic Web-Based Software Structural Testing Using an Adaptive Particle Swarm Optimization Algorithm for Test Data Generation

Abstract

The purpose of a software test is to search for a set of test data in a search space to satisfy a specific coverage criterion. Therefore, finding an effective way to automatically generate this data is an important issue in software testing. This is especially crucial for web-based software, where the size of the program is large, and automatic test-case generation is of prominence. In this paper, a novel method of particle swarm optimization algorithm (PSO) for automatic generation of test data is presented, for web-based software. PSO algorithm has several weaknesses. In this algorithm, there is a possibility of particles to be trapped in local optima. Although PSO is quite rapid compared to other evolutionary algorithms, it usually cannot improve the quality of the solution achieved by increasing iterations. One reason is that in this algorithm, particles converge to a specific point between the best general position and the best personal position. Due to this weakness, a change in PSO has been given in this paper. This is an inertial weight change. In general, in this paper, the inertia weight is dynamically calculated in each round of the algorithm according to the fitness of each particle . Experiments have been performed on different programs and the results of experiments have shown that the proposed method (AIWPSO) has better convergence rate than several methods performed by other variants of the PSO.