Abstract—One of the major expenses about creating and developing a software, is the testing expenses; of which, one of the most important methods is the Mutation Testing. In this method, several versions of the original program including various errors are created and using the appropriate test samples, it’s tried to discover the mistaken versions. The more proper the program and samples, the more number of mistaken versions will be discovered. In order to simulate this method, we need to design a system that executes the test data samples on many mistaken version and then through comparing these results with the main program, discover the versions with error. This method naturally has high calculative and operational costs. The cost includes, repetitive execution of each test data, creating versions with errors and modification of the test samples and mistaken versions if necessary. There’s a general method to reduce the costs. The first category includes the methods which rely on performance expense reduction or reduction of the number of mistaken version. The second rely on fast and cheap data samples of the test that leads to reduction of comparisons between the output of the main program and mistaken versions. In this article, we’re to make a comparison between the methods and introduce the system of mutation tests in order to improve the test data according to evolutionary algorithms.
Index Terms—Mutation testing, genetics algorithm, bacteriological algorithm, mutation testing's cost.
The authors are with form Lahijan Branch, Islamic Azad University, Lahijan, Iran
Cite: Mohsen Falah Rad and Mohadeseh Moosavi, "Investigation of Improving Test Data in Mutation Testing by Optimization Methods," International Journal of Modeling and Optimization vol. 1, no. 5, pp. 438-443, 2011.
Copyright © 2008-2024. International Journal of Modeling and Optimization. All rights reserved.
E-mail: ijmo@iacsitp.com