Software test data generation method based on clustering and multi-population genetic algorithm

Abstract

The invention discloses a software test data generation method based on clustering and a multi-population genetic algorithm, and aims to apply a fuzzy clustering method and a genetic algorithm to software testing to improve the software defect detection efficiency. The method comprises the following steps: firstly, calculating the similarity between variants and the killing difficulty of the variants on the basis of a statistical analyte method under a weak variation test criterion, and sorting the variants; then, based on the sorted variant sequence, selecting the variants which are difficult to kill as a clustering center, and fuzzy clustering the variants; then, for each cluster, establishing a branch coverage constraint-based test data generation mathematical model; and finally, for a plurality of variant clusters, generating test data by adopting a multi-population genetic algorithm based on a strong variation test criterion, preferentially generating variants for killing a clustering center for the variants in each cluster, and dynamically adjusting the clustering center.

Description

technical field [0001] The invention relates to the field of computer software testing, in particular to a method for generating software testing data based on clustering and multi-population genetic algorithms. Background technique [0002] Software quality evaluation is an important research content in the field of software engineering. The correctness and reliability of software need to be evaluated through software testing. Mutation testing is a defect-oriented testing technology proposed by Hamlet and DeMillo. It can not only simulate various types of defects in real software according to the characteristics of programs or statements, but also target specific problems based on the complexity of programs. Select the location where the defect occurs and the number of implanted defects. Therefore, for software testing, mutation testing is a convenient, flexible and personalized technique. Compared with traditional structured testing, mutation testing has adequacy testing...

AI Technical Summary

Problems solved by technology

[0008] In order to solve the problems of low efficiency of mutation testing and low quality of test data in the above-mentioned prior art, the present invention provides a method for generating software test data based on clustering and multi-population genetic algorithms, and deeply excavates the formation mechanism of variants and the relationship between them. Intrinsic correlation, drawing on the idea of ​​"divide and conquer", based on the weak mutation test criterion, adopts the fuzzy clustering method to "divide" the variant; adopts the multi-population genetic algorithm, based on the strong mutation test criterion, "cures" the variant in parallel, and expects to Low test cost, efficient generation of high-quality test data

Images