摘要
包装器是一种能够从网页中自动抽取数据并将其转换为结构化数据的软件程序。现有的包装器生成系统多是半自动化的,需要用户具有关于目标页面的先验知识,而且大多只能处理简单结构数据,而不能很好地处理具有嵌套结构的数据。提出了一种基于后缀树的包装器自动生成方法,生成的包装器不仅可以处理简单结构数据,还可以处理嵌套结构数据,具有较低的时间复杂度,有一定的实用价值。
Wrappers are specialized program routines that automatically extract data from Web pages and convert the information into a structured format.Currently,most approaches to wrapper construction are semi-automated：they either need human involve- ment or have mainly focused on extracting plain-structured data objects with a fixed number of attributes and values,and usally cannot handle nested-structured data objects,whose instances may have variable number of values on their attributes.In this paper,makes a research on a suffix tree based automatic wrapper handle not only data objects with plain structures,but also those and has certain useful value. generation method.The wrapper generated by this method can with nested structures.This method decreases time complexity,
出处
《计算机工程与应用》
CSCD
北大核心
2007年第34期114-118,共5页
Computer Engineering and Applications
基金
国家自然科学基金(the National Natural Science Foundation of China under Grant No.60473042)。
关键词
网页
信息抽取
后缀树
半结构化数据
包装器自动生成
Web page
information extraction
suffix tree
semi-structure data
automatic wrapper generation
