Year: 2018
Type: article
Abstract: The software testing is one of the most predominant and major phase of software development life cycle ensuring the verification and validation process of the software.Test cases should have capability to cover more test objective features i.e., achieve more test coverage.The distinct method for automatic generation of test cases find trivial set of test cases decrease labors as well as budget of software testing and produce maximum effective and operating testing of software product.The most critical step in software development process is the generation of test data.Many methods have been established by researchers to automate it.In this paper we investigate that how P3PGA algorithm improves the quality of the test data and make it more efficient to cover the maximum paths Keywords: Test Data, Test Case, GA, P3PGA I. INTRODUCTION Software testing is an authoritative and self-motivated part of the software development process [1].Software testing is very man power consuming and expensive process [2].The testing efforts are classified in to several parts: test case generation, test execution, test evaluation [1].Test case generation is the fundamental part of any testing process and automating it save copious time effort as well as losses the no of incorrectness and mistakes.An exactly created test suit may not only uncover the faults in a software system but also helps in dropping the high cost, efforts associate with software testing [3].Software testing plays an important role in high quality software development.It practices the application of artificial intelligence techniques.Soft computing is an emerging methodology to computing, whose purpose is to exploit easiness for inaccuracy, indistinctness and partial truth to achieve robustness, submissiveness and total low cost [1].Soft computing are the techniques with in computer science to find out the results to computationally -hard tasks such as the solution of NP-hard problems for which a precise solution should not be derived in polynomial time.There be present a number of techniques for producing test cases like fuzzy logic, finite state machine, soft computing, genetic programing and evolutionary computation [4] .There are two techniques which are used for automated test data generation such as functional and Structural testing.Structural testing are further divided into sub categories which includes different kind of approaches and implementation methods such as Random approach, Path-oriented approach, Goal-oriented approach, Static method, Dynamic method, Hybrid method, Intelligent Approach [5].There are some techniques of test case generation that be determined by on the application like test case generation for web application, object oriented application, structured based system, UML applications etc. 1) UML diagrams: from system requirements we can derive test cases and we can also derive test cases from use cases [6].Some use case models are used for production of test cases.Use case Diagram Graph (UDG) is a diagram for originating test case that demonstrates all the possible case in the SUT.2) Critical Path Method: T.Y.Chen et al [7] projected a framework for producing test cases i.e. critical path method.By using this technique we generate test cases by purifying functional choices.3) Code based test Generation: [A.Mathur] test cases can be produced directly from the code which is called as code based test generation.This method decreases the size of the test suite, or prioritizes tests by supporting regression testing.4) Dynamic Path Testing and Evolutionary Technique: it is dynamic path testing technique that engenders test case by implementing the program with different probable test case values.Korel also offered a methodology of test case prioritization established on path testing.With the help of random testing test cases are organized according to the smallest path [8].J. Wegener and H.Sthamer generate test cases with the help of several structural test coverage testing criteria by evolutionary approaches like Genetic algorithm, which frequently generate test cases by using Meta heuristic method.
Author: Nargis Akhter
Language: English
Subfield: Software
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W2864208861