欢迎您！东篱公司
退出
-
申报数据库
-
立项数据库
-
成果数据库
-
项目获奖数据库
中文摘要：
针对事件抽取获得的单个元事件无法完整描述主题事件的特点,提出了一种主题事件的融合方法,通过该方法将与同一主题相关的所有元事件整合在一起,以层次化的形式表示。首先定义了一种事件融合框架TEFF（topic event fusion framework）。该框架根据各类元事件在主题事件中的作用,将主题事件以层次化的形式表示。同时给出元事件和主题的相关度计算方法,通过该算法来评价元事件和主题的相关度。在TEFF的指导下,通过相关度计算,实现主题事件的融合。在以2008年起的金融危机为主题的实验中,取得了F值为77.1%的实验结果,这表明该方法能有效地对主题事件进行融合。
英文摘要：
A single event description cannot describe a topic event completely. This paper proposed a topic event fusion method to fuse correlative events which describe the same topic and represent in hierarchy form. First, defined topic event fusion framework （TEFF）. The framework could organize the topic with the form of hierarchy by analyzed the events＇ role in the topic. And proposed the correlation algorithm of events and topic. Under the direction of TEFF, and thorough the correlation computing, could achieve the topic event fusion. The capability of fusing correlative events is testified by the experiment of ＂financial crisis＂ topic fusion and show that the F-measure achieves 77.1%. It proved that the method takes a good effect in topic event fusion.
同期刊论文项目
同项目期刊论文
Exploring syntactic structured features over parse trees for relation extraction using kernel method
Label propagation via bootstrapped support vectors for semantic relation extraction between named en
期刊信息
位置：