Method for selecting regression test case for clustering with semi-supervised information

Abstract

The invention discloses a method for selecting a regression test case for clustering with semi-supervised information. The method comprises the following steps: recording the execution overage information of the test case, generating a function execution profile, and representing the test case in a quantitative form; analyzing the historical test results to obtain the constraint relationship among test cases; and analyzing the test cases with a semi-supervised clustering algorithm to obtain similarities and differences of the execution conditions of the test cases, understand the relation between program behaviors and the test cases, effectively reduce the number of test cases in the regression test stage and maintaining enough high error detection capability. According to the invention, the program is understood according to the internal relation of the program behaviors revealed by the test cases based on the data mining technology so that the selection of the test cases is easier and more automatic, the tests cases can be used more effectively in regression tests, the test case selection accuracy is promoted, and the regression test efficiency is improved.

Description

technical field [0001] The invention belongs to the technical field of computer software testing, and relates to a test case selection technology in regression testing and a program behavior clustering analysis technology in software testing, which are used to improve the efficiency of regression testing and enable test cases to have higher error detection capabilities, specifically A method for regression test case selection using semi-supervised information for clustering. Background technique [0002] As the software version changes, the program is constantly modified. During the modification process, new bugs may be introduced into the original functions of the program. The regression test technology reuses the previously developed test case set to test the original program. function, so as to confirm that the correctness of the original function of the program has not been affected by the modification. However, since the original test case set is usually very large, ex...

AI Technical Summary

Problems solved by technology

[0012] The technical problem to be solved by the present invention is: due to the lack of full use of the historical test results obtained in the test process and the test experience of the tester itself, the existing regression test case selection technology is on the balance of the number of test cases and the error detection ability Poor performance; propose a method using semi-supervised cluster analysis in regression test case selection techniques, through a deep understanding of program behavior, deduce from historical test results or given by the tester itself between test cases Constraint information is used to improve clustering results to effectively reduce the number of test cases and maintain a sufficiently high error detection capability

Images