Generating test data covering multiple paths using genetic algorithm incorporating with reducing input domain

Abstract

The problem of efficiently generating test data covering multiple paths was focused on this study, and a method of generating test data covering multiple paths using a genetic algorithm incorporating with reducing the input domain of a program was presented. In this method, all target paths are first divided into several groups based on the same independent sub-path, and the input variables corresponding to the independent sub-path are determined. Then, a multi-population genetic algorithm is used to generate test data to cover these target paths, each sub-population generating test data covering target paths belonging to the same group. During the evolution, the input variables corresponding to the traversed independent sub-path are remained fixed, and the ranges of crossover and mutation operations are reduced, leading to these sub-populations’ search in a reduced input domain so that the efficiency of generating test data is improved.