摘要
Web表格作为一种简洁有效的数据信息表达方式,已广泛应用于Web页面中。现提出一种基于表格结构的Web表格信息抽取模型,该模型主要有表格定位模块、表格结构预处理模块和表格信息抽取与重构模块三个模块组成,根据Web表格的结构标记和自定义的启发式规则来抽取表格信息。实验结果表明该模型能够很好地应用于Web表格信息的抽取。
As a compact and efficient way to present relational data information, Web tables are used frequently in Web documents. In this paper it presents a new model based on table structure that extracts information from tables of Web documents. It is composed of table positio- ning 'module, table structure pretreatment module and table information extraction and remodelling module, extracts information from table according to Web table structure label and self-defined heuristic rules. The experimental results show that this model is well performed in information extraction from tables of Web documents.
出处
《计算机应用与软件》
CSCD
2009年第4期72-74,共3页
Computer Applications and Software
基金
国家发改委基金项目(SNMCFIP-2006S001)。
关键词
表格结构
抽取模型
启发式规则
预处理
解析
Table structure Model of extracting Heuristic method rules Pretreatment Parse