Naive Bayes

How It Works and Why It’s Useful

Simple & Powerful

Naive Bayes

A probabilistic classifier based on Bayes' theorem. Assumes features are independent.

What is it?

P(A|B) = [P(B|A) * P(A)] / P(B).  Calculates probability of A given B.

Bayes' Theorem

Features are independent. Simplifies calculations, hence "naive."

Naive Assumption

Gaussian (continuous data), Multinomial (text), Bernoulli (binary)

Types

Spam filtering, text classification, medical diagnosis, and more!

Applications

Easy to implement, fast, works well with high-dimensional data.

Why Use It?