Comparative Evaluation of the State-of-art Requirements-based Test Case Generation Approaches

Abstract

The overall aim of software testing is to deliver the error-free and high-quality software products to the end users. The testing process ensures that a software is aligned with the user specification and requirements. In software testing process, there are many challenging tasks however test case generation process is considered as the most challenging one. The quality of the generated test cases has a significant impact on efficiency and effectiveness of the testing process. In order to improve the quality of a developed software, the test cases should be able to achieve maximum adequacy in the testing and requirements' coverage. This paper presents a comparative evaluation of the prominent requirement-based test case generation approaches. Five evaluation criteria namely, inputs for test case generation, transformation techniques, coverage criteria, time and tool's support are defined to systematically compare the approaches. The results of the evaluation are used to identify the gap in the current approaches and research opportunities in requirements-based test case's generation.

An Ontology for Specifying and Tracing Requirements Engineering Artifacts and Test Artifacts

“…Several authors have proposed strategies for the derivation of test cases based on requirements artifacts [1], [10], [11]. While some of them raise the possibility of deriving test cases directly from the artifacts (such as the System Requirement Specification document, use cases diagrams, or textual use cases) [2], others propose test case derivation from the so-called "test models" (model-based test-case derivation approach) [12], [13].…”

“…In the process of developing a software or system, use cases represent a means to specify the functional requirements that have to be implemented, in a more detailed and concrete way. Use cases can be also used to drive the generation of the test cases [1], [10] that will be executed to verify that the software meets the requirements.…”

“…When defining a test case, the corresponding interactions can be defined by means of scenarios at different levels of abstraction. Pohl defines in [10] the concept of test case scenario (TestCaseScenario) to refer to "types of test cases", which describe sets of test cases that have certain types of inputs and certain types of expected outputs (TestInputType and TestExpectedOutputType, respectively).…”

“…A recent work [10] reports the state of the art about strategies for the generation of test cases from requirements artifacts. That paper analyzes the literature on the automatic derivation of test cases, classifying it according to various criteria, like what type of input device is used, whether they use intermediate test models or not, the transformation techniques applied, and the degree of coverage considered.…”

Improved Test Case Selection Algorithm to Reduce Time in Regression Testing

“…1 the third phase (C) the assembling segment calculation of time selection material from the running test cases collection version. To improve the efficiency, and reduce redundancy, time and unnecessary recurrences as well as fault detected of already executed and passed test cases need depth analysis of modified test suite [10][11][12][13][14]. There should be a focus on mentioned areas as much as possible.…”