Study And Design Of Information Integration Model Based On Web Pages Content

The information integration technology has developed greatly for 20 years. Researchers have proposed a number of integration architectures and methods. However, most of their attention focuses on heterogeneous databases, and their aims are often to make access different well-structured relation databases seamlessly. With the development of Internet, especially the web technologies, an amount of web pages increasingly well up. These semi-structured and unstructured web data form a huge information database. Without certain structured pattern, clear semantic information and high efficient access application programming interface, these incompact information only can be shown, explained and recognized by browsers. It is very difficult to reuse these data frequently as relational data. However, there is too much information that always affects people's life in that huge database. Therefore, how to integrate and make full use of these web resources has begun to attract many researchers' attention.Firstly, the integration development statuses in quo of semi-structured and unstructured data are summarized in this thesis. Then the characters of the web information structure are analyzed. To probe into methods of web information, after the author's considering the traditional information integration methods, making full use of the developing technology such as Java, xml, web service, DOM, parsing and extraction technology, a method of integrating web information and an integration model framework are proposed, and correspond solutions to the different parts of the model are suggested.