Research On Generation Of Test Case Based On UML And Intelligent Optimization Algorithm

With the rapid development of computer science and technology, and promote the widespread use of application software products. Whether producers or users of the software,all seeking to survive in the fierce competition and development, vital quality of software products, and has become the focus of attention in order to ensure the quality of software products, software testing has become essential Important processes and methods. In the development process of software systems in the face of extremely complex issues, software developers can not fully comply with the subjective understanding of the objective facts, but also to communicate with the project closely related to various types of personnel between the impossible and the perfect fit. Thus, at every stage of the software life cycle will inevitably produce an error. Although sought through a rigorous technical review, as found before the end of each stage very early and correct errors, however, not possible to examine all the errors found, and, in the process of correcting these errors may also introduce new errors. If you do not find in the software put in before the actual operation and correct it for errors, these errors will be exposed during the run up to the time not only to correct these errors and costly, and the system is likely to have serious consequences.This paper introduces the basic concepts and methods of software testing test case generation of the common techniques are reviewed and analyzed and discussed a variety of technology has advantages and disadvantages; Secondly, test method based on UML made simple set, UML models are used to guide pointed out the advantages of testing lies, and analyzes the UML activity diagram testability and test strategies, and further noted that the activity diagram model is not only a powerful tool for business needs analysis and system design, but also an important basis for system testing. Then study the ant colony algorithm and its improvements, gives the corresponding improved algorithms, making the application of improved ant colony algorithm when traversing a directed graph, the resulting set of paths required for testing, and not just of an optimal path.Finally, this paper presents a test case generation system design, analysis and flow of its modules, and the online reservation system as an example to explain the process in detail using this method to generate an abstract test case. The results showed that the use of UML activity diagram models and improved ant colony algorithm to generate test cases, test cases can not only generate redundancy effectively controlled to generate a reasonable number of test cases, test cases and generate effective to meet test requirements.The results of this study:1) the analysis of testability and benefits on the basis of activity diagrams, UML activity diagram presented converted into a directed graph of rules and algorithms.2) On the basis of the standard ant colony algorithm, an improved ant colony algorithm is applicable to test of the generated path, including changing the way ants choose to walk the path, the definition of artificial ants property, that energy is infinite, memory body, death methods.