| As the components of InternetWare Software on the internet are more and more, an effective management platform, component library system is very important.But the existing component library system has many problems such as poor component,poor description resources,single query.To solve these problems, this article will work on three aspects.First on the component resource collection, in this paper, based on the search engine and combining the specific website crawler oriented method, and the collection methods is verified by the experiments of high availability, and then define and describe a support for a variety of classification of component model.Next in the arrangement of component information, this article focuses on the component information related to the validation work, put forward using text mining technology to choose the high correlation between the description of the information.Finally on the retrieval of component, this paper improve the text query methods, put forward a intelligent query condition extension method, and the experiment show that this method is compared with the traditional did not extend the query conditions is more effective. At the same time this paper also supports browsing query mode, makes user retrieval more convenient. |
