知识卡片
姓氏语料类别不同于个人国籍
教学数据混合语言、地区和国家来源标签,只能描述该语料如何收集和分类姓氏。共享姓氏、迁移及多种语言使用使标签并非个人身份真值;不能据姓名推断现实人物国籍、族群或宗教。
这张卡片看不懂?
这张卡片已通过人工审查,不再接受失败反馈。
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
8.2 Predicting nationality with only a last name To get you up to speed quickly on recycling, you’ll start with the simplest possible token: the lowly character. You will build a model that can predict the nationality of last names, using only the letters in the names to guide the predictions. This kind of model may not sound all that useful to you. You might
参考来源
- 位置:《Natural Language Processing in Action, Second Edition》第08章《循环网络与长短期记忆》(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-8.html)
- 结论依据:摘录中关于姓氏语料类别不同于个人国籍的定义、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:8.2 Predicting nationality with only a last name To get you up to speed quickly on recycling, you’ll start with the simplest possible token: the lowly character. You will build a model that can predict the nationality of last names, using only the letters in the names to guide the predictions. This kind of model may not sound all that useful to you. You might