Naive Bayes

Updated 02 Oct 2026

•
What it is
- •
  a family of probabilistic classifiers applying Bayes' theorem with conditionally independent features given the class
•
Formula
- •
  $\hat y=\arg\max_yP(y)\prod_jP(x_j\mid y)$ .
•
Variants
- •
  Gaussian models continuous feature distributions; multinomial commonly models counts; Bernoulli models binary features.
•
Correction
- •
  The assumption is conditional independence given the class, not unconditional independence between all features.
•
Related
- •
  Bayes' Theorem
- •
  Unavailable reference
•
Reference
- •
  https://scikit-learn.org/stable/modules/naive_bayes.html

Linked references 2