知识卡片

普通读书笔记卡 · 9251.b · 生成模型:GPT-6

朴素贝叶斯不要求特征同等重要

理解等级:中级

类内条件独立允许联合似然乘积分解,各特征对不同类别的似然比可以不同,因此贡献并不必须相等。原文混写P(A|B)和P(B|A)时应按贝叶斯公式统一条件方向。

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 fundamental Naïve Bayes assumption is that each feature is independent and has an equal contribution. Example: The following example demonstrates the working of this algorithm. Suppose you are trying to decide whether to go on a hike. The decision to go hiking will be based on the weather condition. We are interested in predicting the chance or probability that the decision to hike is

    参考来源

    - 位置:《Machine Learning Fundamentals: Concepts, Models, and Applications》第06章《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/19_chapter06.xhtml) - 结论依据:摘录中关于朴素贝叶斯不要求特征同等重要的定义、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 fundamental Naïve Bayes assumption is that each feature is independent and has an equal contribution. Example: The following example demonstrates the working of this algorithm. Suppose you are trying to decide whether to go on a hike. The decision to go hiking will be based on the weather condition. We are interested in predicting the chance or probability that the decision to hike is - 生成时纠错核验:[官方机制定义](https://scikit-learn.org/stable/modules/naive_bayes.html)(2026-11-06;非代码执行或独立内容审查)。