Research On Ant Colony Algorithm Based Component Retrieval Method | | Posted on:2010-12-20 | Degree:Master | Type:Thesis | | Country:China | Candidate:T Huang | Full Text:PDF | | GTID:2178360272479340 | Subject:Computer application technology | | Abstract/Summary: | PDF Full Text Request | | The software development method based on component reuse is one of the available ways to resolve the crisis of software. In the process of the software development based on componet, how to retrieve and extract the most satisfying component in the library of component is important in component library management system. The retrieval efficiency of the component library not only includes the retrieval of the component itself, but also includes the understanding of the component. However, the general component retrieval system provides descriptive information for understanding the component, but seldom provides information about respect of reused actually. Data mining technology provides a feasible method for the problem.In the paper, how to use the application of ant colony algorithm to mine rules of component reuse is discussed. Firstly, the theory and technology of component reuse and data mining are discussed. Especially, the theory of component retrieval and the ant colony classification algorithm are discussed. Secondly, based on the work, data mining technology is used to discover the rule of component reuse. The rule provides auxiliary decision support to component reusers. The component retrieval method based on ant colony classification algorithm is proposed in order to resolve the problems that the reuse rules are acquired. And an applied model of software component retrival based on ant colony algorithm is created. Lastly, the model has been verified through experiment, so the feasibility and validity of this strategy has been proved. | | Keywords/Search Tags: | Software, Software Reuse, Component Library, Component Retrieval, Data Mining, Ant Colony Algorithm, Classification Rule | PDF Full Text Request | Related items |
| |
|