论文部分内容阅读
随着W eb的发展,越来越多的数据可以通过表单提交来获取,这些表单提交所产生信息是由D eep W eb后台数据库动态产生的。在这种情况下,信息集成就更加需要W eb爬虫来自动获取这些页面以进一步地处理数据。为了帮助用户完成这样的任务,提出一种用于搜集D eep W eb页面的爬虫的设计方法。此方法使用一个预定义的领域本体知识库来识别这些页面的内容,同时利用一些来自W eb站点的导航模式来识别自动填写表单时所需进行的路径导航。通过对来自不同领域的D eep W eb站点的大量实验,验证了此方法是非常有效的。
As Web grows, more and more data is available through form submissions, and the information submitted by these forms is dynamically generated by the Data Web back-end database. In this case, W eb crawlers are even more needed for information integration to automatically acquire these pages to process the data further. To help users accomplish this task, a design methodology for crawlers that collect D ebbs Web pages is proposed. This method uses a predefined domain ontology repository to identify the contents of these pages and utilizes some navigation patterns from the Web site to identify the path navigation needed to auto-complete the form. Through a large number of experiments from D eep W eb sites in different fields, this method is proved to be very effective.