Software Engineering

A. V. K. Shanthi ¹, G. Mohan Kumar ²

Affiliations
1 Sathyabama University, Chennai, IN
2 Park College of Engineering, Coimbatore, IN

Abstract

Software testing is an important part of software development life cycle. As part of automating testing process, automated test case generation is presented in this paper. Test case generation from design specifications has the advantage of generating test cases before software development cycle. In our approach, Evolutionary Genetic algorithm is used to generate all possible valid test cases for the software to be tested or developed. Data is obtained from UML design i.e., class diagram. A Novel method is proposed to form tree structure among class in design, based on their relationships. Depth First Searching (DFS) algorithm is used to generate test cases from the binary tree formed. All requirements of data mining concept implemented in very established way to produce optimal number test cases. A case study that is going to uses a simple, but extensively studied program is used to illustrate the approach.

Keywords

Depth First Searching, Data Mining, Genetic Algorithm, Testing, Test Cases, and UML (Unified Modeling Language).

Full Text

     

A. V. K. Shanthi ¹, D. Parthiban ¹, G. Mohan Kumar ²

Affiliations
1 Sathyabama University, Chennai, IN
2 Park Engineering College, Coimbatore, IN

Abstract

Software testing is an important phase that ensures the quality of the software. Where test cases are an important entity or criteria by which software is being evaluated. A test case is being generated by various approaches but by means of Modeling Languages attracts the recent researches and industrialists. This paper focus on test case generation by means of UML Class diagram. Novel method is proposed to form tree structure among class in design based on their relationships. From formed tree Genetic Two point Cross over is applied to obtain all possible new trees. Depth First Searching (DFS) algorithm is used to generate test cases from the binary tree formed. Problems in usage of certain techniques are identified. A case study that is going to uses a simple, but extensively studied program is used to illustrate our approach.

Keywords

Software Testing, Test Cases, UML (Unified Modeling Language), Genetic Crossover Operator, DFS (Depth First Search) and Class Diagram.

Full Text

     

A. V. K. Shanthi ¹, G. Mohan Kumar ²

Affiliations
1 Sathyabama University, Chennai, IN
2 Park Engineering College, Coimbatore, IN

Abstract

Recent studies have shown that neglected conditions are common and can have dramatic consequences to the operations of a Software defects. Neglected conditions in software can compromise the accuracy of a software product or even cause global disruptions to the quality of a Product. Several solutions have been proposed. However, these solutions share a common limitation: they are based on rules which need to be known beforehand. Violations of these rules are deemed to the problem of detecting the neglected conditions. As policies typically differ among different software’s, these approaches are limited in the scope of mistakes they can detect. In this paper, we address the problem of detecting neglected conditions in software using data mining technique. We apply association rules mining to the program files of the software product across an administrative domain to discover local, software product-specific policies. Deviations from these local policies are potential for detecting the neglected conditions. In this evaluation, we focused on three aspects of the configurations: syntax, interfaces and function calls. Here the advanced data mining techniques is used to discover implicit conditional rules in a code base and to discover rule violations that indicate neglected conditions. More interestingly, our empirical study indicates that our approaches are able to discover ordering rules, which involve ordering of function calls, and the corresponding rule violations also.

Keywords

Neglected Conditions, Data Mining, Software Defects.

Full Text