Dissertation > Excellent graduate degree dissertation topics show

Fitness Function Design for Nested If-Else and Function Call Flow Involved Construct in Evolutionary Testing

Author: ZhangMiao
Tutor: LiuXiYang
School: Xi'an University of Electronic Science and Technology
Course: Computer Software and Theory
Keywords: Software Testing Automated test data generation Evolution test Fitness function
CLC: TP311.52
Type: Master's thesis
Year: 2008
Downloads: 22
Quote: 1
Read: Download Dissertation

Abstract


Evolutionary testing is very important to the rise in recent years of an automated test data generation techniques. The main idea of ??the evolutionary test is the test data generation problem is transformed into the evolutionary search problem, in order to achieve the automation of the test. Evolution of the global search capability of testing the use of evolutionary algorithms, under the guidance of the oriented fitness function, able to quickly and efficiently generate test data automatically. The fitness function is a very important aspect of the evolutionary test. A well-designed fitness function to provide better and more accurate guide to the evolutionary search, thus improving the efficiency of evolutionary search. Therefore, the fitness function has been evolutionary hot research field testing. Many well-designed fitness function is applied to evolutionary search. Nested if-else structure, due to its special multi-layer nested structure, when the test data in a layer of deviation from the target branch, other branches nested in the layer will no longer be performed at this time directly calculate the appropriate value. This situation has led the loss of information of the satisfaction degree of the test data in the inner layer is not performed on the branch. Resulting in unfair evaluation of the test data. To address this issue, we draw on the idea of ??a measurable transformation, proposed the concept of branch optimism. Accumulated branch distance calculation on the test data in the execution of the branch by branch optimism. Added by branch optimism normalized to the original fitness function, a fitness function for nested if-else structure is successfully raised. When the execution of the test objectives related to the structure of function calls, the existing fitness function is no longer applicable. This is due to the existing static analysis of the fitness function only for test target located inside a function, and does not consider test target dependence of the function call. This involves a function call test target, this article by function calls, static analysis, highlighting key branch in the chain of function calls identified methods, proposed the concept of function approximation degree. Function approximation degrees used to measure test data dependency chain to meet the test objectives in the function call. Ultimately, a new fitness function involved in the function call test target. Two fitness function proved through the experimental test, nested If-Else structure and function call structure to more effectively guide evolutionary search to improve search efficiency.

Related Dissertations

  1. The Application of Evolutionary Testing to Nested Branch Structure of Assembly Language in Embedded Software,TP368.1
  2. The Automatic Generation of MM Paths Based on the Polymorphism of Object-Oriented Software,TP311.52
  3. Embedded software testing technology - oriented research,TP311.52
  4. Research on the Testing Method of the Web Application Based on Combined Open Source Frameworks,TP311.52
  5. Research on Object-Oriented Integration Testing Method,TP311.53
  6. Automated management process - oriented application software testing Bug Management System Design and Implementation,TP311.52
  7. The Design and Implementation of NL Software Test Case Library Management System,TP311.52
  8. The Research and Application on CMMI-Based SaaS Software Test Process Improvement,TP311.52
  9. A Study on Fitness Calculation Problem in Evolutionary Testing of C Program,TP311.53
  10. Research on the Scheme for Testing Object-Oriented Software,TP311.52
  11. Research on Method and Tool for the Missile Mother Vehicle Embedded Software Testing,TP311.52
  12. A Method for Evaluating the Security of VoIP Services,TN916.2
  13. The Design and Implementation of An Electronic Exam Management System,TP311.52
  14. The Research on Run-time Software Test,TP311.53
  15. Research of System Security Verification Technology Combined with Testing,TP311.52
  16. Based on simulated annealing algorithm automatically generate test data EFSM model,TP311.52
  17. Based on Rough Set Neural Network software testing outsourcing costs Prediction Model,TP311.53
  18. Design and Implementation of Distibuted and Automatic Availability Evaluation Platform Based on STAF,TP311.52
  19. Research on Test Data Generation Based on Simulated Annealing Genetic Algorithm,TP311.52
  20. Flight instructor comprehensive quality assessment system design and implementation,TP311.52
  21. Categories Based E-commerce Navigation System Design and Implementation,TP311.52

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer software > Program design,software engineering > Software Engineering > Software Development
© 2012 www.DissertationTopic.Net  Mobile