我试图从恐怖事件的文本数据中提取一些特性来创建一个数据集。使用名称实体识别,我已经成功地提取了名称、位置、组织等特征,现在我想提取与事件相关的成员数量。
The 2008 Mumbai attacks (also referred to as 26/11) were a series of terrorist attacks that took place in
November 2008, when 10 members of Lashkar-e-Taiba, a terrorist organization based in Pakistan,
carried out 12 coordin
因此,我正在使用ANTLR语法来解析日期,我希望不仅能够识别单个的日期单位,而且能够识别成对的日期单位。
就这个问题而言,我认为把我想要识别的问题分为三类也许会有帮助:
What was the temperature in August 2019? -直截了当。单一日期-单位(8月2019).Which was hotter between June 3rd 2019 and yesterday? -仍然直截了当。两个日期-单位(2019年6月3日和yesterday).Between August 2018 and 2019, which was hotter? -棘手。在这种情况下,用户