Web表格信息抽取的研究

Web表格信息抽取是信息抽取在Web表格上的一种应用，是当今的一个研究热点。本文首先分析了Web表格信息抽取的过程，包括表格识别、结构识别以及“属性-值”对的提取；然后对当前国内外在基于特定域和独立城两种表格信息抽取研究方法上的动态及成果追行了比较和分析。在此基础上，提出了表格抽取的关键技术——表格结构识别上的一些想法；最后展望了Web表格信息抽取技术的发展趋势。

Research on Web Table Extraction

Web table extraction, which is a current research hotspot, is an application of information extraction on Web table. In this paper, we first analyze the flow of Web table extraction, including table detection, structure recognition and attribute-value pair extraction. Then we compare what others have done with both domain-specific and domain-independent methodologies ir this field. Based on the above survey and analysis, we put forward some ideas in the table structure recognition, which is one of the key steps in the flow of whole extraction. At last, we present the tendency of development of Web table extraction.