运用文本处理框架抽取中文事件

Study on Chinese event extraction based on GATE

针对目前网页文档中文事件抽取的关键问题,该文提出利用开源的通用文本处理框架进行中文事件抽取的方法。设计了文本处理框架中文事件处理流程,开发文本处理框架插件,解决中文分词与词性标注、领域词典、中文抽取规则设计等关键技术,实现了中文事件的类型识别和元素抽取。并以4类政治事件为例,进行中文事件抽取实验。实验结果表明,基于文本处理框架的中文事件抽取具有良好的通用性及较好的抽取效果。

Event extraction is one of the most important research field in information extrac- tion. Aiming at the key problem of Chinese event extraction in the web page document, a method of Chi- nese event extraction with general architecture for text engineering （GATE） was proposed. The procedure of GATE Chinese event was designed, several GATE plug-in were developed to solve key technologies of Chinese word segmentation and part of speech tagging, domain dictionary and Chinese extraction rule de sign. This paper took five category political events extraction for instance, made an events extraction ex periment. The result showed that Chinese event extraction method based on GATE could be applied uni- versally and had good result.