Test Data Generation Method Of Artificial Bee Colony Based On Huffman Coding

With the rapid improvement of information technology and the popularity of mobile terminals,the demand for various types of software has also increased dramatically.Software quality has gradually become the focus of attention.Software testing is an important task to ensure the quality of software during software development,and it is also an important measure to improve software reliability and ease of use.Test cases determine the quality of software testing.How to design effective test cases,how to improve software test performance is a key issue in software testing.A large number of studies have shown that search-based software testing is one of the effective methods to improve test performance.This method uses heuristic search technology to generate test data,and heuristic search technology is used to evaluate a set of solutions in the fitness function search space.In practice,the program under test is usually intricate,and the number of target paths of the program under test can reach an infinite number.It is not feasible to use the exhaustive test.Therefore,many scholars have explored a large number of methods based on heuristic search technology for software testing,such as white box structure testing,black box functional testing,gray box structure and functional combination testing.However,the existing research results need to improve the performance of test case generation.Therefore,this paper proposes a path coverage test data generation method based on Huffman coded artificial bee colony.The artificial bee colony is a new pseudo-biological group algorithm,which has the characteristics of fast convergence,less control parameters and strong robustness,and has obvious advantages in optimizing complex problems.In this paper,the artificial bee colony algorithm is applied to test data generation,which not only can improve the working efficiency of testers,but also improve the efficiency of test data generation,reduce development costs and maximize economic benefits.Firstly,this paper proposes a test data generation method based on artificial bee colony algorithm for single path coverage.Compared with the single path coverage method of genetic algorithm,the experimental results show that the artificial bee colony algorithm has more advantages in test performance.Then,a multipath coverage method based on Huffman coding-artificial bee colony algorithm is proposed.First,Huffman coding is performed on the target path,and a new objective function(fitness function)is designed according to the coding.Secondly,a path coverage test data generation system model based on Huffman coded artificial bee colony algorithm is proposed and designed.The model consists of three modules: program analysis,algorithm execution and test drive.Finally,in order to better verify the practicability and efficiency of this model,the proposed method is used in the test of triangle classification program,three-number sorting and bubble sorting benchmark program,and similar methods(MMPPSO,Ahmed method and The single path method)is used for comparison.The experimental results show that the path coverage test data generation method based on Huffman coding-artificial bee colony algorithm has obvious advantages in terms of path coverage and running time.The artificial bee colony algorithm is more adaptive and efficient.