摘要： 随着互联网的普及，信息技术的发展，形成了大量的新闻信息资源。从海量的新闻信息中抽取出有用的资源，是当前迫切需要解决的问题。该文在分析新闻网页结构的基础上，结合了基于DOM 的结构抽取和基于文本特征模式抽取两种处理技术的优点，提出了基于Web 新闻网页的半自动化抽取技术，自动下载了有用的Web 页面，抽取了所需的新闻信息。最后，该文描述了一个面向奥运新闻的信息抽取系统，并给出了该系统的实验结果。

关键词: 信息抽取；包装器；DOM；抽取规则

Abstract: With the widespread use of Internet and the development of information technology, there are a tremendous amount of news information resource. The ability to quickly obtain useful resource from the huge news information is a crucial problem at present. Based on the analysis of news information, this paper introduces an approach of semi-automatically extracting from Web resource. Moreover, it gives the system which extracts useful Olympic news information and experiment results of it.

Key words: Information extraction; Wrapper; DOM; Extraction rule