摘要
为了从具有海量信息的Internet上自动抽取Web页面的信息,提出了一种基于树比较的Web页面主题信息抽取方法。通过目标页面与其相似页面所构建的树之间的比较,简化了目标页面,并在此基础上生成抽取规则,完成了页面主题信息的抽取。对国内主要的一些网站页面进行的抽取检测表明,该方法可以准确、有效地抽取Web页面的主题信息。
In order to automatically extract Web page information from Internet that contains magnanimous information, this paper presented an approach based on tree comparison. This approach compared tree built from the target page with that ones built from its similar pages to simplify the target page. Extraction rules were generated on this basis, and then we used the rules to extract topic information from the target Web page. Experiment result shows this extraction method is precise and efficient.
出处
《微型机与应用》
2011年第19期67-69,共3页
Microcomputer & Its Applications
基金
上海市科委国际合作项目(10510712500)
关键词
信息抽取
相似页面
树比较
抽取规则
information extraction
similar pages
tree comparison
extraction rules
