关键词: 本体, Web, 信息抽取

Abstract: Aiming at the malpractice in traditional search field based on keyword and data retrieval, this paper proposes a Web information extraction framework based on ontology. Firstly, the framework obtains the Web page which is converted into a wellformed HTML document, secondly, the document is turned into the DOM tree by making use of the HTML parser, then, the extraction rules is achieved on the basis of the users’ interest data block which is obtained according to the XPath expression. Finally, the data is extracted through the OntPMatch algorithm, and is stored in RDF data format. The paper makes the empirical study using the cotton information as research object, and realizes a prototype system of extracting biological information data. The paper provides a useful tool for users to obtain valuable biological information from Web.

Key words: ontology, Web, information extraction