Informatics Publishing Limited

A. Jalila ¹, D. Jeya Mala ¹

Affiliations
1 Department of Computer Applications, Thiagarajar College of Engineering, IN

Abstract

The adoption of fault detection techniques during initial stages of software development life cycle urges to improve reliability of a software product. Specification-based testing is one of the major criterions to detect faults in the requirement specification or design of a software system. However, due to the non-availability of implementation details, test case generation from formal specifications become a challenging task. As a novel approach, the proposed work presents a methodology to generate test cases from OCL (Object constraint Language) formal specification using Category Partitioning Method (CPM). The experiment results indicate that the proposed methodology is more effective in revealing specification based faults. Furthermore, it has been observed that OCL and CPM form an excellent combination for performing functional testing at the earliest to improve software quality with reduced cost.

Keywords

Software Testing, Specification-Based Testing, OCL, CPM, Formal Specification.

Full Text

     

R. Uma Maheswari ¹, D. Jeya Mala ²

Affiliations
1 Department of Computer Applications, K.L.N. College of Engineering, Sivagangai - 630612, Tamil Nadu, IN
2 Department of Computer Applications, Thiagarajar College of Engineering, Madurai - 625005, Tamil Nadu, IN

Abstract

Background: The test case prioritization of regression Testing is described. Methods: A new test case prioritization algorithm is proposed to get better the rate of fault detection and cost reduction. Heuristic Techniques like Genetic Algorithm (GA) and Simulated Annealing (SA) are employed. It prioritizes the test cases depends on fault detection ability and execution time taken. Findings: The implementation of proposed algorithm in JAVA is found to produce optimal or near optimal results. The efficiency of proposed regression testing technique is proved by comparing it with GA and SA methods individually. Applications: A complete automation tool for the complete usage of the algorithm is being developed. It will also be analyzed on larger projects with large number of test cases and faults.

Keywords

Fault Coverage, Genetic Algorithm, Heuristic Techniques, Regression Testing, Simulated Annealing, Test Case Prioritization

Full Text

   

B. Geetha ¹, D. Jeya Mala ²

Affiliations
1 Department of Computer Science and Engineering, Anna University, Chennai, IN
2 Department of Master of Computer Applications, Fatima college, IN

Abstract

All research made on the Object-Oriented (OO) paradigms focus on the fundamentals of analysis, programming, and design. The primary problem found in testing the systems which are object-oriented is a methodology of standard testing and this may not be very useful. The test case may execute software using a new set consisting of some input values and will then compare them to the output to check if the test has passed. An optimum test case set is obtained using a process of selection that is viewed to be a problem of optimization. Thus, metaheuristic optimizing or searching is a technique used often for optimizing or searching which is used in automated testing of software. The BAT Algorithm is a metaheuristic that is dependent on the property of echolocation of the miniaturized scale bats. The property further controls the conduct of search of the bats of a small-scale and making them discover prey thus enabling them to identify distinctive types of bugs irrespective of the fact they are found to be dull. The work also proposed a new and hybrid Tabu search algorithm using the BAT for the selection of test case.

Keywords

Object Oriented (OO) Paradigms, Hybrid Tabu Search, Bat Algorithm, Test Case Selection.

Full Text