Naive Bayes

NAIVE BAYES applies Bayes' theorem assuming features are INDEPENDENT — 'naive' because they usually aren't.

Surprisingly effective for text classification (spam, sentiment).

Example 1 (python)

from sklearn.naive_bayes import MultinomialNB
m = MultinomialNB().fit(X_train, y_train)
print(m.score(X_test, y_test))

Output

0.87

Very fast to train.

Example 2 (python)

# GaussianNB for continuous, MultinomialNB for counts, BernoulliNB for binary

Pick the variant to match your data.

Key points

💡 Note: Multinomial Naive Bayes on TF-IDF features is a strong, cheap baseline for spam and sentiment tasks.