STM Journals Peer-reviewed research from Consortium e-Learning Network

Journal of Advanced Database Management & Systems

A Framework for Test Data Generation Using Genetic Algorithm

  1. Bindhyachal Kumar Singh
  2. Arun Solanki
  3. Ashish Kumar

Abstract

Software testing is one of the best methods to increase the confidence of the programmers in terms of correctness, quality and reliability of software. The testing effort is divided into three parts: test case generation, test execution, and test assessment. The goal of software testing is to design a set of minimal number of test data such that it reveals as many faults as possible. This paper proposed a framework for generating efficient test data for path testing using genetic algorithm (GA). The proposed framework accepts a program unit files and then generate a control flow diagram (CFG). The CFG is used for identification of paths in a detailed form with the help of path selection criteria. The identified path is used to select a target path using probability theory. Finally the selected path and execution of genetic algorithm are used to generate the final test data. The proposed framework also provides a facility for evaluation of generated test data to ensure their effectiveness in terms of coverage and efficiency. The proposed framework reduces time, effort and also increases the quality of generated test data. This paper is organized into three parts: part I discuss the applicability of genetic algorithm in software testing, part II describes the proposed framework, part III presents the experimental case study.Cite this ArticleBindhyachal Kumar Singh, Arun Solanki, Ashish Kumar. A Framework for Test Data Generation Using Genetic Algorithm. Journal of Advanced Database Management & Systems. 2016; 3(1): 18–25p.

Keywords

Support