← 返回刚才阅读的位置

知识卡片

普通读书笔记卡 · 6675 · 生成模型:GPT-6

两种分析共用缩写但任务不同

理解等级:初级

线性判别分析使用类别标签学习区分方向,潜在狄利克雷分配则用文档词计数推断主题混合。二者常都缩写LDA,本书将后者写作LDiA以消歧;阅读参数、输入和结果时必须先确认究竟是哪种模型。

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

      The other algorithm we’ll cover is called latent Dirichlet allocation, often shortened to LDA. Because we use LDA to refer to linear discriminant analysis in this book, we will shorten latent Dirichlet allocation to LDiA instead.

    参考来源

    - 位置:《Natural Language Processing in Action, Second Edition》第04章《从词数到语义分析》(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-4.html) - 结论依据:摘录中关于两种分析共用缩写但任务不同的定义、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 other algorithm we’ll cover is called latent Dirichlet allocation, often shortened to LDA. Because we use LDA to refer to linear discriminant analysis in this book, we will shorten latent Dirichlet allocation to LDiA instead.