知识卡片

普通读书笔记卡 · 6660 · 生成模型:GPT-6

中文分词需区分字词和语法标记

理解等级:中级

汉语词边界常不以空格明确标出,一个字可独立成词或参与多字词,分词规范还会影响粒度。一座中的座是量词,不表示古老;们、在、过等承担各自语法功能,不能机械当英语复数或过去时词尾等价物。

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

      Chinese, Japanese, and other logographic languages aren’t limited to a small number of letters in alphabets used to compose tokens or words. Characters in these languages look more like drawings and are called logographs. There are many thousands of unique characters in the Chinese language, and these characters are used much like we use words in alphabet-based languages, such as English. But each Chinese character

    参考来源

    - 位置:《Natural Language Processing in Action, Second Edition》第02章《思维的词元》(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-2.html) - 结论依据:摘录中关于中文分词需区分字词和语法标记的定义、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:Chinese, Japanese, and other logographic languages aren’t limited to a small number of letters in alphabets used to compose tokens or words. Characters in these languages look more like drawings and are called logographs. There are many thousands of unique characters in the Chinese language, and these characters are used much like we use words in alphabet-based languages, such as English. But each Chinese character