Generating software test data by evolution

This paper discusses the use of genetic algorithms (GAs) for automatic soft ware test data generation. This research extends previous work on dynamic t est data generation where the problem of test data generation is reduced to one of minimizing a function [1], [2]. In our work, the function is minimi zed by using one of two genetic algorithms in place of the local minimizati on techniques used in earlier research. We describe the implementation of o ur GA-based system and examine the effectiveness of this approach on a numb er of programs, one of which is significantly larger than those for which r esults have previously been reported in the literature. We also examine the effect of program complexity on the test data generation problem by execut ing our system on a number of synthetic programs that have varying complexi ties.