Research Article
Research on Software Test Data Optimization Using Adaptive Differential Evolution Algorithm
@INPROCEEDINGS{10.1007/978-3-031-50549-2_17, author={Zheheng Liang and Wuqiang Shen and Chaosheng Yao}, title={Research on Software Test Data Optimization Using Adaptive Differential Evolution Algorithm}, proceedings={Advanced Hybrid Information Processing. 7th EAI International Conference, ADHIP 2023, Harbin, China, September 22-24, 2023, Proceedings, Part III}, proceedings_a={ADHIP PART 3}, year={2024}, month={3}, keywords={Adaptive differential evolution algorithm Data selection Individual clustering Fitness function Software test data Optimization}, doi={10.1007/978-3-031-50549-2_17} }- Zheheng Liang
Wuqiang Shen
Chaosheng Yao
Year: 2024
Research on Software Test Data Optimization Using Adaptive Differential Evolution Algorithm
ADHIP PART 3
Springer
DOI: 10.1007/978-3-031-50549-2_17
Abstract
In order to improve the coverage of the target path corresponding to the generated software test data after optimization, and make the data better adapt to the software test process, the adaptive differential evolution algorithm is introduced to design the optimization method for software test data. Using the PSO algorithm to simulate the biological evolution mechanism in nature, and with the help of computer programming, the generated software test data are preliminarily trained; Draw on the path correlation based regression test data evolution of adaptive differential evolution algorithm to generate the relevant path representation method, mark the program to be tested, and construct the test data fitness function based on this; A hybrid model MPSO is proposed to select the best individual data in software test data; The selection criteria of scaling individuals are introduced into the adaptive scaling factor to cluster the optimal data individuals, so as to realize the design of optimization methods. The comparison experiment results show that the designed method has a good effect in practical application. This method can improve the target path coverage corresponding to the generated software test data on the basis of controlling the time length and evolution times required for software test data optimization.
