知识卡片

普通读书笔记卡 · 6772.c · 生成模型:GPT-6

实体标签与词性不必串行依赖

理解等级:中级

词性、依存和实体注释有助不同任务,但NER不必先读取tagger或parser输出才工作。部分组件共享词向量或Transformer表示,应按实际依赖保留组件;机械删除共享表示可能影响多个后续任务。

这张卡片看不懂?

已有反馈 · 0 条

    还没有人提交反馈。

    在整书笔记中的位置

    原书出处 · 1 处

    • · 查看原书依据

      The challenge of NER is closely related to a more basic problem: POS tagging. To recognize named entities in the sentence, you need to know which POS each word belongs to. As discussed earlier, in English grammar, the POS for a person, place, or thing is noun, and your named entity will often be a proper noun—a noun that refers to a particular person, place

    参考来源

    - 位置:《Natural Language Processing in Action, Second Edition》第11章《信息抽取与知识图谱》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Natural Language Processing in Action, Second Edition/_epub-src/OEBPS/Text/chapter-11.html) - 结论依据:摘录中关于实体标签与词性不必串行依赖的定义、比较或条件支撑卡片;原书表述不严谨的范围已在覆盖记录逐项限定,不把案例结果外推为普遍结论。 - 原始内容:The challenge of NER is closely related to a more basic problem: POS tagging. To recognize named entities in the sentence, you need to know which POS each word belongs to. As discussed earlier, in English grammar, the POS for a person, place, or thing is noun, and your named entity will often be a proper noun—a noun that refers to a particular person, place