Study On The Application Of Swarm Intelligence Algorithm For Test Case Set Optimization

Software testing is an essential part of the software development procedure.In order to check whether the developed software system meets the original requirements,the software system is normally tested manually or automatically.The test case set optimization technique is to pre-process the original test case set to significantly reduce the cost of software testing and speed up software development.There are many techniques for optimizing test case sets in software testing.In this paper,we focus on two of the common test case set optimization techniques.Test suite reduction is to remove the duplicate or invalid test cases that are added during the iterative process of versioning,so as to reduce the cost loss of software testing,but still ensure that all test points can be covered,thus ensuring the quality of software testing.In this paper,we improve the ant lion optimizer with diverse populations and high optimality seeking and introduce it into the study of test suite reduction problem.The program under test of the smart meter system is selected as the data set in this paper.The experimental results show that,compared with the commonly used test suite reduction algorithms,the proposed ant lion optimizer can better remove the duplicate or invalid test cases from the original test case set,and can quickly obtain the optimal test case subset.Test case prioritization is to reorder the sequence of test cases required for software testing,so that the test cases which can better cover the test points in the original test case set are ranked in the front for execution,thus covering all the test points in the shortest time to achieve the software testing requirements.In this paper,we propose a test case prioritization algorithm based on the artificial fish school algorithm,which is designed to optimize the three intelligent behaviors of the artificial fish school.The experimental results show that,compared with the commonly used intelligent optimization algorithms,the artificial fish school algorithm proposed in this paper can reorder the original test case set better in both single-and multi-objective aspects,thus improving the software testing efficiency.