基于DOM树的非规范化表格信息定位技术

Location Technology of Non-standardized Table Based on DOM Tree

Web表格信息提取已经成为构建本体的重要内容之一,它能自动将本体所需的属性名和属性值提取出来,节省大量人工劳动。关于非规范化表格信息提取的研究比较少,对本体构建造成大量信息缺失。提供一种基于启发式规则的非规范化表格信息定位算法,其对定位非规范化表格准确率较高。

The information extraction of web table has become the important task of construct ontology. It extracts attrib- ute name and value for ontology automatically so that large volume human task can be saved. There are few studies for in- formation extraction of non-standardized table in the domestic and overseas. The above phenomenon causes information- missing in the process of building ontology. The present paper proposed a heuristic and inerratic location algorithm of non- standardized table which can provide a much higher accuracy rate for locating informal table.