知识卡片

普通读书笔记卡 · 9255.e · 生成模型:GPT-6

大方差方向不保证类别可分

理解等级:中级

PCA按无标签变异而非类别分离度选方向。类别信息可能在低方差方向,大方差也可能来自无关噪声;降维后是否有助分类必须用独立下游评价,不凭解释方差自动下结论。

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

      PCA works by considering the variance of each feature as the high attribute variance shows a good split between the classes, and hence it reduces the dimensionality. Some real-world applications of PCA are image processing, movie recommendation systems, and optimizing the power allocation in various communication channels.

    参考来源

    - 位置:《Machine Learning Fundamentals: Concepts, Models, and Applications》第07章《无监督学习模型》(来源项目:a19c7933-4eda-4176-aebd-30fe7e1ad292;源文件:extracted/Machine Learning Fundamentals: Concepts, Models, and Applications/_epub-src/OEBPS/xhtml/20_chapter07.xhtml) - 结论依据:摘录中关于大方差方向不保证类别可分的定义、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:PCA works by considering the variance of each feature as the high attribute variance shows a good split between the classes, and hence it reduces the dimensionality. Some real-world applications of PCA are image processing, movie recommendation systems, and optimizing the power allocation in various communication channels.