← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 9243.b · 生成模型:GPT-6

类别编码不自动保留业务次序

理解等级:中级

名义类别无固有顺序,序数类别有领域次序。按字母排序映成整数不会自动得到cut质量等级,也不保证相邻数值代表等距离;编码应匹配变量语义和下游模型。

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

      In a data set, it is not unusual to find categorical features. The data set with categorical or qualitative variables are not numbers and they are classified as nominal or ordinal. The nominal data are the data that are classified in a category where there is no order involved. An example of such data is the color of your car or marital status. Since there

    参考来源

    - 位置:《Machine Learning Fundamentals: Concepts, Models, and Applications》第05章《使用数据》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Machine Learning Fundamentals: Concepts, Models, and Applications/_epub-src/OEBPS/xhtml/17_chapter05.xhtml) - 结论依据:摘录中关于类别编码不自动保留业务次序的定义、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:In a data set, it is not unusual to find categorical features. The data set with categorical or qualitative variables are not numbers and they are classified as nominal or ordinal. The nominal data are the data that are classified in a category where there is no order involved. An example of such data is the color of your car or marital status. Since there