Automated Test Data Generation Based on a Genetic Algorithm with Maximum Code Coverage and Population Diversity

In the present paper, we investigate an approach to intelligent support of the software white-box testing process based on an evolutionary paradigm. As a part of this approach, we solve the urgent problem of automated generation of the optimal set of test data that provides maximum statement coverage of the code when it is used in the testing process. We propose the formulation of a fitness function containing two terms, and, accordingly, two versions for implementing genetic algorithms (GA). The first term of the fitness function is responsible for the complexity of the code statements executed on the path generated by the current individual test case (current set of statements). The second term formulates the maximum possible difference between the current set of statements and the set of statements covered by the remaining test cases in the population. Using only the first term does not make it possible to obtain 100 percent statement coverage by generated test cases in one population, and therefore implies repeated launch of the GA with changed weights of the code statements which requires recompiling the code under the test. By using both terms of the proposed fitness function, we obtain maximum statement coverage and population diversity in one launch of the GA. Optimal relation between the two terms of fitness function was obtained for two very different programs under testing.

Search-based Test Data Generation for Mutation Testing: a tool for Python programs

Test data generation for mutation testing consists of identifying a set of inputs that maximizes the number of mutants killed. Mutation Testing is an excellent test criterion for detecting faults and measuring the effectiveness of test data sets. However, it is not widely used in practice due to the cost and complexity to perform some activities as generating test data. Although test suites can be produced and selected manually by a tester this practice is susceptible to errors and tools are needed to facilitate it. Several tools have been developed to automate mutation testing, but, only a few address the test data generation. The present paper proposes an automated test data generation tool based on weak mutation for Python programming language using the Hill Climbing algorithm. For evaluation, we performed an experiment concerning the effectiveness and cost computational of the tool in a database composed of 348 mutants and we compare it with random generation. Overall, the experiment achieved an average mutation score of 86% for our proposed tool and random testing 64% on average.

An Analysis on the Applicability of Meta-Heuristic Searching Techniques for Automated Test Data Generation in Automatic Programming Assessment

Automatic Programming Assessment (APA) has been gaining lots of attention among researchers mainly to support automated grading and marking of students’ programming assignments or exercises systematically. APA is commonly identified as a method that can enhance accuracy, efficiency and consistency as well as providing instant feedback on students’ programming solutions. In achieving APA, test data generation process is very important so as to perform a dynamic testing on students’ assignment. In software testing field, many researches that focus on test data generation have demonstrated the successful of adoption of Meta-Heuristic Search Techniques (MHST) so as to enhance the procedure of deriving adequate test data for efficient testing. Nonetheless, thus far the researches on APA have not yet usefully exploited the techniques accordingly to include a better quality program testing coverage. Therefore, this study has conducted a comparative evaluation to identify any applicable MHST to support efficient Automated Test Data Generation (ATDG) in executing a dynamic-functional testing in APA. Several recent MHST are included in the comparative evaluation combining both the local and global search algorithms ranging from the year of 2000 until 2018. Result of this study suggests that the hybridization of Cuckoo Search with Tabu Search and lévy flight as one of promising MHST to be applied, as it’s outperforms other MHST with regards to number of iterations and range of inputs.