Particle swarm based evolution and generation of test data using mutation testing

Abstract

Adequate test data generation is a vital task involved in the process software testing. Process of mutation testing, a fault-based testing technique, generates mutants of the program under test (PUT) by applying mutation operators. These mutants can assist in finding test cases that have the potential to detect faults in the PUT. Particle Swarm Optimisation (PSO) share similar working characteristics with Genetic Algorithm (GA) which has already been applied to test data generation using mutation testing. In this paper, applicability of PSO for the generation of test data with mutation testing is explored. The results obtained by empirical evaluation of the proposed approach on benchmark C programs are presented. The evaluated results show that the test cases generated from the technique proposed kills substantial number of mutants and therefore, has a scope of exploring its performance in the area of search based test case generation.