Dissertation > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer software > Program design,software engineering > Software Engineering > Software Development

Research and Implementation of the new software testing technology

Author XiongJiao
Tutor LuoGuangChun
School University of Electronic Science and Technology
Course Applied Computer Technology
Keywords Automatic Test Data Generation Path - oriented test Genetic Algorithms The use of Markov chain model
CLC TP311.52
Type Master's thesis
Year 2008
Downloads 367
Quotes 0
Download Dissertation

Software testing is the primary means to ensure software quality and reliability, the design of the test data is an important part of software testing, the automatic generation of test data is critical and difficult. With the development of software technology, the size of the program gradually increases, the complexity is also increased gradually. In the software development process, completely relying on human the analysis test efficiency is too low, but can not guarantee the quality of software. A major problem in the testing process is to generate the test data with a certain degree of coverage, and these data do not belong to the equivalence classes. If you have a tool that can automatically analyze program and generate test data, will greatly improve the reliability of the software and save a lot of manpower. There are a lot of tools automatically generate test data, but most have some limitations, can not be completely automatically generate test data that can be handled by the data type is also limited and can only be used for the local unit test. This paper begins with a brief introduction in the field of research progress on the basic theory of automated software focus in-depth study of the software testing heuristics genetic algorithm and Markov chain model, and improved on this basis are given a new test case design methods and how under the automatic generation of test cases. Analyzed on the basis of the genetic algorithm adaptive genetic algorithm improved adaptive genetic algorithm and simulation analysis. While the introduction of genetic algorithms and simulated annealing algorithm combined genetic simulated annealing algorithm to verify the superiority of the genetic simulated annealing algorithm in search optimization problem. The use of genetic simulated annealing algorithm to automatically generate a path-oriented test case data, and test cases based on coverage. The use of Markov chain model is a random process model to describe the use of the software, the first software statistical test methods based on Markov chain model is constructed software use Markov chain model to describe the use of the software, and combined with the genetic The simulated annealing algorithm to automatically generate the state transition probability matrix, is used to guide the generation of test cases. Finally the use of genetic algorithms and Markov chain model combined Hi-Tech Software Testing prototype system of automated testing tools.

Related Dissertations
More Dissertations