Automatic generation of test data for path testing by adaptive genetic simulated annealing algorithm

Abstract

Software testing has become an important stage of the software developing process in recent years, and it is crucial element of software quality assurance. Path testing has become one of the most important unit test methods, and it is a typical white box test. The generation of testing data is one of the key steps which have a great effect on the automation of software testing. GA is adaptive heuristic search algorithm premised on the evolutionary ideas of natural selection and genetic. Because it is a robust search method requiring little information to search effectively in a large or poorly-understood search space, it is widely used to search and optimize, and also can be used to generate test data. In this article we put the anneal mechanism of the Simulated Anneal Algorithm into the genetic algorithm to decide to accept the new individuals or not, and we import dynamic selections to adaptive select individuals which can be copied to next generation. Adaptive crossover probability, adaptive mutation probability and elitist preservation ensure that the best individuals can not be destroyed. The experiment results show that adaptive genetic simulated annealing algorithm is superior to genetic algorithm in effectiveness and efficiency.