句子分解 丁文韬.

Presentation on theme: "句子分解 丁文韬."— Presentation transcript:

1 句子分解 丁文韬

2 句子分解 流程 原则 基于syntax tree把句子拆分成若干部分 识别拆出的部分之间的关系，构造新的树
拆分出得到的部分应该是更细致的理解的输入 拆分出的每个部分应该能表达充足的语义

3 拆分句子的方式 拆分依据（syntax tree结构） 拆分得到的每个部分都必须包含下列的结构之一
完整陈述 S | (WH-word | NP)…VP 省略主语的陈述 CC … VP | to VP | V+“ing” … 单独的项 (“,”|“and”|“or”) NP (“,”|“.”|“;”) 括号内的内容 “-LRB-” … “-RRB-”

4 拆分方法 首先从后向前检测所有单独的项，将其挂在前 方的NP上 从上至下BFS 切下子树时：根节点归在最左子树所属的一半
从前向后（左边界小的优先）枚举连续的一段文本 对应的树结构是否满足模式 若满足，检查切下这一段后剩余部分是否满足模式 若均满足，执行拆分，终止遍历并对得到两个新的 结果调用拆分算法 切下子树时：根节点归在最左子树所属的一半

5 示例1 Carl_Friedrich_Gauss
A contested story relates that, when he was eight, he figured out how to add up all the numbers from 1 to 100. A contested story relates that, when he was eight, he figured out how to add up all the numbers from 1 to 100.

6 示例1 Carl_Friedrich_Gauss
when he was eight, he figured out how to add up all the numbers from 1 to 100 when he was eight ,he figured out how to add up all the numbers from 1 to 100

7 示例2 Mass_media For example, it is controversial whether to include cell phones , computer games (such as MMORPGs ), and video games in the definition.

8 示例2 Mass_media For example, it is controversial whether to include {cell phones, computer games, video games in the definition}. For example, it is controversial whether to include …

9 Q & A Thanks for listening